feat(pageserver): use vectored_get in collect_keyspace

Signed-off-by: Alex Chi Z <chi@neon.tech>
storcon: signal LSN wait to pageserver during live migration (#10452 )
2026-07-03 20:20:38 +00:00 · 2025-01-28 20:31:15 +01:00 · 2025-01-28 17:33:07 +00:00 · 2025-01-28 17:21:05 +00:00 · 2025-01-28 17:08:17 +00:00 · 2025-01-28 15:32:59 +00:00
133 changed files with 4968 additions and 1894 deletions
--- a/.github/actions/neon-project-create/action.yml
+++ b/.github/actions/neon-project-create/action.yml
@@ -17,6 +17,31 @@ inputs:
  compute_units:
    description: '[Min, Max] compute units'
    default: '[1, 1]'
+  # settings below only needed if you want the project to be sharded from the beginning
+  shard_split_project:
+    description: 'by default new projects are not shard-split, specify true to shard-split'
+    required: false
+    default: 'false'
+  admin_api_key:
+    description: 'Admin API Key needed for shard-splitting. Must be specified if shard_split_project is true'
+    required: false
+  shard_count:
+    description: 'Number of shards to split the project into, only applies if shard_split_project is true'
+    required: false
+    default: '8'
+  stripe_size:
+    description: 'Stripe size, optional, in 8kiB pages.  e.g. set 2048 for 16MB stripes. Default is 128 MiB, only applies if shard_split_project is true'
+    required: false
+    default: '32768'
+  psql_path:
+    description: 'Path to psql binary - it is caller responsibility to provision the psql binary'
+    required: false
+    default: '/tmp/neon/pg_install/v16/bin/psql'
+  libpq_lib_path:
+    description: 'Path to directory containing libpq library - it is caller responsibility to provision the libpq library'
+    required: false
+    default: '/tmp/neon/pg_install/v16/lib'
+  

 outputs:
  dsn:
@@ -63,6 +88,23 @@ runs:
        echo "project_id=${project_id}" >> $GITHUB_OUTPUT

        echo "Project ${project_id} has been created"
+
+        if [ "${SHARD_SPLIT_PROJECT}" = "true" ]; then
+          # determine tenant ID
+          TENANT_ID=`${PSQL} ${dsn} -t -A -c "SHOW neon.tenant_id"`
+          
+          echo "Splitting project ${project_id} with tenant_id ${TENANT_ID} into $((SHARD_COUNT)) shards with stripe size $((STRIPE_SIZE))"
+
+          echo "Sending PUT request to https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split"
+          echo "with body {\"new_shard_count\": $((SHARD_COUNT)), \"new_stripe_size\": $((STRIPE_SIZE))}"
+          
+          # we need an ADMIN API KEY to invoke storage controller API for shard splitting (bash -u above checks that the variable is set)
+          curl -X PUT \
+            "https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split" \
+            -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
+            -d "{\"new_shard_count\": $SHARD_COUNT, \"new_stripe_size\": $STRIPE_SIZE}"
+        fi
+
      env:
        API_HOST: ${{ inputs.api_host }}
        API_KEY: ${{ inputs.api_key }}
@@ -70,3 +112,9 @@ runs:
        POSTGRES_VERSION: ${{ inputs.postgres_version }}
        MIN_CU: ${{ fromJSON(inputs.compute_units)[0] }}
        MAX_CU: ${{ fromJSON(inputs.compute_units)[1] }}
+        SHARD_SPLIT_PROJECT: ${{ inputs.shard_split_project }}
+        ADMIN_API_KEY: ${{ inputs.admin_api_key }}
+        SHARD_COUNT: ${{ inputs.shard_count }}
+        STRIPE_SIZE: ${{ inputs.stripe_size }}
+        PSQL: ${{ inputs.psql_path }}
+        LD_LIBRARY_PATH: ${{ inputs.libpq_lib_path }}
--- a/.github/workflows/_build-and-test-locally.yml
+++ b/.github/workflows/_build-and-test-locally.yml
@@ -158,8 +158,6 @@ jobs:

      - name: Run cargo build
        run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
          ${cov_prefix} mold -run cargo build $CARGO_FLAGS $CARGO_FEATURES --bins --tests

      # Do install *before* running rust tests because they might recompile the
@@ -217,8 +215,6 @@ jobs:
        env:
          NEXTEST_RETRIES: 3
        run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
          LD_LIBRARY_PATH=$(pwd)/pg_install/v17/lib
          export LD_LIBRARY_PATH

@@ -229,8 +225,13 @@ jobs:
          ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E '!package(pageserver)'

          # run pageserver tests with different settings
-          for io_engine in std-fs tokio-epoll-uring ; do
-            NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+          for get_vectored_concurrent_io in sequential sidecar-task; do
+            for io_engine in std-fs tokio-epoll-uring ; do
+              NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO=$get_vectored_concurrent_io \
+                NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine \
+                ${cov_prefix} \
+                cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+            done
          done

          # Run separate tests for real S3
@@ -314,6 +315,7 @@ jobs:
          CHECK_ONDISK_DATA_COMPATIBILITY: nonempty
          BUILD_TAG: ${{ inputs.build-tag }}
          PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
+          PAGESERVER_GET_VECTORED_CONCURRENT_IO: sidecar-task
          USE_LFC: ${{ matrix.lfc_state == 'with-lfc' && 'true' || 'false' }}

      # Temporary disable this step until we figure out why it's so flaky
--- a/.github/workflows/build-macos.yml
+++ b/.github/workflows/build-macos.yml
@@ -235,7 +235,7 @@ jobs:
          echo 'CPPFLAGS=-I/usr/local/opt/openssl@3/include' >> $GITHUB_ENV

      - name: Run cargo build (only for v17)
-        run: PQ_LIB_DIR=$(pwd)/pg_install/v17/lib cargo build --all --release -j$(sysctl -n hw.ncpu)
+        run: cargo build --all --release -j$(sysctl -n hw.ncpu)

      - name: Check that no warnings are produced (only for v17)
        run: ./run_clippy.sh
--- a/.github/workflows/build_and_test.yml
+++ b/.github/workflows/build_and_test.yml
@@ -820,8 +820,8 @@ jobs:
      - name: Print logs and clean up
        if: always()
        run: |
-          docker compose -f ./docker-compose/docker-compose.yml logs || 0
-          docker compose -f ./docker-compose/docker-compose.yml down
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml logs || true
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml down

  promote-images-dev:
    needs: [ check-permissions, tag, vm-compute-node-image, neon-image ]
@@ -859,7 +859,7 @@ jobs:
          done

  promote-images-prod:
-    needs: [ check-permissions, tag, test-images, vm-compute-node-image ]
+    needs: [ check-permissions, tag, test-images, promote-images-dev ]
    runs-on: ubuntu-22.04
    if: github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'

@@ -1078,12 +1078,6 @@ jobs:
              console.log(`Tag ${tag} created successfully.`);
            }

-            // TODO: check how GitHub releases looks for proxy/compute releases and enable them if they're ok
-            if (context.ref !== 'refs/heads/release') {
-              console.log(`GitHub release skipped for ${context.ref}.`);
-              return;
-            }
-
            try {
              const existingRelease = await github.rest.repos.getReleaseByTag({
                owner: context.repo.owner,
@@ -1102,7 +1096,8 @@ jobs:
                owner: context.repo.owner,
                repo: context.repo.repo,
                tag_name: tag,
-                generate_release_notes: true,
+                // TODO: Automate release notes properly
+                generate_release_notes: false,
              });
              console.log(`Release for tag ${tag} created successfully.`);
            }
--- a/.github/workflows/ingest_benchmark.yml
+++ b/.github/workflows/ingest_benchmark.yml
@@ -28,7 +28,24 @@ jobs:
    strategy:
      fail-fast: false # allow other variants to continue even if one fails
      matrix:
-        target_project: [new_empty_project, large_existing_project]
+        include:
+          - target_project: new_empty_project_stripe_size_2048 
+            stripe_size: 2048 # 16 MiB
+            postgres_version: 16
+          - target_project: new_empty_project_stripe_size_32768
+            stripe_size: 32768 # 256 MiB # note that this is different from null because using null will shard_split the project only if it reaches the threshold
+                               # while here it is sharded from the beginning with a shard size of 256 MiB
+            postgres_version: 16
+          - target_project: new_empty_project
+            stripe_size: null # run with neon defaults which will shard split only when reaching the threshold
+            postgres_version: 16
+          - target_project: new_empty_project
+            stripe_size: null # run with neon defaults which will shard split only when reaching the threshold
+            postgres_version: 17
+          - target_project: large_existing_project
+            stripe_size: null # cannot re-shared or choose different stripe size for existing, already sharded project
+            postgres_version: 16
+      max-parallel: 1 # we want to run each stripe size sequentially to be able to compare the results
    permissions:
      contents: write
      statuses: write
@@ -67,17 +84,21 @@ jobs:
        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}

    - name: Create Neon Project
-      if: ${{ matrix.target_project == 'new_empty_project' }}
+      if: ${{ startsWith(matrix.target_project, 'new_empty_project') }}
      id: create-neon-project-ingest-target
      uses: ./.github/actions/neon-project-create
      with:
        region_id: aws-us-east-2
-        postgres_version: 16
+        postgres_version: ${{ matrix.postgres_version }}
        compute_units: '[7, 7]' # we want to test large compute here to avoid compute-side bottleneck
        api_key: ${{ secrets.NEON_STAGING_API_KEY }}
+        shard_split_project: ${{ matrix.stripe_size != null && 'true' || 'false' }}
+        admin_api_key: ${{ secrets.NEON_STAGING_ADMIN_API_KEY }} 
+        shard_count: 8
+        stripe_size: ${{ matrix.stripe_size }}

    - name: Initialize Neon project
-      if: ${{ matrix.target_project == 'new_empty_project' }}
+      if: ${{ startsWith(matrix.target_project, 'new_empty_project') }}
      env:
          BENCHMARK_INGEST_TARGET_CONNSTR: ${{ steps.create-neon-project-ingest-target.outputs.dsn }}
          NEW_PROJECT_ID: ${{ steps.create-neon-project-ingest-target.outputs.project_id }}
@@ -130,7 +151,7 @@ jobs:
        test_selection: performance/test_perf_ingest_using_pgcopydb.py
        run_in_parallel: false
        extra_params: -s -m remote_cluster --timeout 86400 -k test_ingest_performance_using_pgcopydb
-        pg_version: v16
+        pg_version: v${{ matrix.postgres_version }}
        save_perf_report: true
        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
      env:
@@ -146,7 +167,7 @@ jobs:
        ${PSQL} "${BENCHMARK_INGEST_TARGET_CONNSTR}" -c "\dt+"

    - name: Delete Neon Project
-      if: ${{ always() && matrix.target_project == 'new_empty_project' }}
+      if: ${{ always() && startsWith(matrix.target_project, 'new_empty_project') }}
      uses: ./.github/actions/neon-project-delete
      with:
        project_id: ${{ steps.create-neon-project-ingest-target.outputs.project_id }}
--- a/.github/workflows/neon_extra_builds.yml
+++ b/.github/workflows/neon_extra_builds.yml
@@ -114,7 +114,7 @@ jobs:
        run: make walproposer-lib -j$(nproc)

      - name: Produce the build stats
-        run: PQ_LIB_DIR=$(pwd)/pg_install/v17/lib cargo build --all --release --timings -j$(nproc)
+        run: cargo build --all --release --timings -j$(nproc)

      - name: Configure AWS credentials
        uses: aws-actions/configure-aws-credentials@v4
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -3,8 +3,9 @@ name: Create Release Branch
 on:
  schedule:
    # It should be kept in sync with if-condition in jobs
-    - cron: '0 6 * * FRI' # Storage release
    - cron: '0 6 * * THU' # Proxy release
+    - cron: '0 6 * * FRI' # Storage release
+    - cron: '0 7 * * FRI' # Compute release
  workflow_dispatch:
    inputs:
      create-storage-release-branch:
@@ -55,7 +56,7 @@ jobs:
      ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}

  create-compute-release-branch:
-    if: inputs.create-compute-release-branch
+    if: ${{ github.event.schedule == '0 7 * * FRI' || inputs.create-compute-release-branch }}

    permissions:
      contents: write
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -179,7 +179,7 @@ dependencies = [
 "nom",
 "num-traits",
 "rusticata-macros",
- "thiserror",
+ "thiserror 1.0.69",
 "time",
 ]

@@ -718,14 +718,14 @@ dependencies = [

 [[package]]
 name = "axum"
-version = "0.7.9"
+version = "0.8.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "edca88bc138befd0323b20752846e6587272d3b03b0343c8ea28a6f819e6e71f"
+checksum = "6d6fd624c75e18b3b4c6b9caf42b1afe24437daaee904069137d8bab077be8b8"
 dependencies = [
- "async-trait",
 "axum-core",
 "base64 0.22.1",
 "bytes",
+ "form_urlencoded",
 "futures-util",
 "http 1.1.0",
 "http-body 1.0.0",
@@ -733,7 +733,7 @@ dependencies = [
 "hyper 1.4.1",
 "hyper-util",
 "itoa",
- "matchit 0.7.0",
+ "matchit",
 "memchr",
 "mime",
 "percent-encoding",
@@ -746,7 +746,7 @@ dependencies = [
 "sha1",
 "sync_wrapper 1.0.1",
 "tokio",
- "tokio-tungstenite 0.24.0",
+ "tokio-tungstenite 0.26.1",
 "tower 0.5.2",
 "tower-layer",
 "tower-service",
@@ -755,11 +755,10 @@ dependencies = [

 [[package]]
 name = "axum-core"
-version = "0.4.5"
+version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "09f2bd6146b97ae3359fa0cc6d6b376d9539582c7b4220f041a33ec24c226199"
+checksum = "df1362f362fd16024ae199c1970ce98f9661bf5ef94b9808fee734bc3698b733"
 dependencies = [
- "async-trait",
 "bytes",
 "futures-util",
 "http 1.1.0",
@@ -942,6 +941,18 @@ version = "1.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8c3c1a368f70d6cf7302d78f8f7093da241fb8e8807c05cc9e51a125895a6d5b"

+[[package]]
+name = "bb8"
+version = "0.8.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d89aabfae550a5c44b43ab941844ffcd2e993cb6900b342debf59e9ea74acdb8"
+dependencies = [
+ "async-trait",
+ "futures-util",
+ "parking_lot 0.12.1",
+ "tokio",
+]
+
 [[package]]
 name = "bcder"
 version = "0.7.4"
@@ -1118,7 +1129,7 @@ dependencies = [
 "log",
 "nix 0.25.1",
 "regex",
- "thiserror",
+ "thiserror 1.0.69",
 ]

 [[package]]
@@ -1299,9 +1310,9 @@ dependencies = [
 "serde_with",
 "signal-hook",
 "tar",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-stream",
 "tokio-util",
 "tower 0.5.2",
@@ -1408,9 +1419,9 @@ dependencies = [
 "serde",
 "serde_json",
 "storage_broker",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-util",
 "toml",
 "toml_edit",
@@ -1786,11 +1797,24 @@ dependencies = [
 "chrono",
 "diesel_derives",
 "itoa",
- "pq-sys",
- "r2d2",
 "serde_json",
 ]

+[[package]]
+name = "diesel-async"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "51a307ac00f7c23f526a04a77761a0519b9f0eb2838ebf5b905a58580095bdcb"
+dependencies = [
+ "async-trait",
+ "bb8",
+ "diesel",
+ "futures-util",
+ "scoped-futures",
+ "tokio",
+ "tokio-postgres 0.7.12",
+]
+
 [[package]]
 name = "diesel_derives"
 version = "2.2.1"
@@ -2239,7 +2263,7 @@ dependencies = [
 "pin-project",
 "rand 0.8.5",
 "sha1",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-util",
 ]
@@ -3365,12 +3389,6 @@ dependencies = [
 "regex-automata 0.1.10",
 ]

-[[package]]
-name = "matchit"
-version = "0.7.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b87248edafb776e59e6ee64a79086f65890d3510f2c656c000bf2a7e8a0aea40"
-
 [[package]]
 name = "matchit"
 version = "0.8.4"
@@ -3761,7 +3779,7 @@ dependencies = [
 "serde_json",
 "serde_path_to_error",
 "sha2",
- "thiserror",
+ "thiserror 1.0.69",
 "url",
 ]

@@ -3811,7 +3829,7 @@ dependencies = [
 "futures-sink",
 "js-sys",
 "pin-project-lite",
- "thiserror",
+ "thiserror 1.0.69",
 "tracing",
 ]

@@ -3843,7 +3861,7 @@ dependencies = [
 "opentelemetry_sdk",
 "prost",
 "reqwest",
- "thiserror",
+ "thiserror 1.0.69",
 ]

 [[package]]
@@ -3879,7 +3897,7 @@ dependencies = [
 "percent-encoding",
 "rand 0.8.5",
 "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-stream",
 "tracing",
@@ -3993,7 +4011,7 @@ dependencies = [
 "remote_storage",
 "serde_json",
 "svg_fmt",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-util",
 "utils",
@@ -4042,8 +4060,8 @@ dependencies = [
 "pageserver_compaction",
 "pin-project-lite",
 "postgres",
- "postgres-protocol",
- "postgres-types",
+ "postgres-protocol 0.6.4",
+ "postgres-types 0.2.4",
 "postgres_backend",
 "postgres_connection",
 "postgres_ffi",
@@ -4069,12 +4087,12 @@ dependencies = [
 "strum_macros",
 "sysinfo",
 "tenant_size_model",
- "thiserror",
+ "thiserror 1.0.69",
 "tikv-jemallocator",
 "tokio",
 "tokio-epoll-uring",
 "tokio-io-timeout",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-stream",
 "tokio-tar",
 "tokio-util",
@@ -4115,7 +4133,7 @@ dependencies = [
 "storage_broker",
 "strum",
 "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
 "utils",
 ]

@@ -4130,9 +4148,9 @@ dependencies = [
 "postgres",
 "reqwest",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-stream",
 "tokio-util",
 "utils",
@@ -4438,7 +4456,7 @@ dependencies = [
 "futures-util",
 "log",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 ]

 [[package]]
@@ -4459,6 +4477,24 @@ dependencies = [
 "stringprep",
 ]

+[[package]]
+name = "postgres-protocol"
+version = "0.6.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "acda0ebdebc28befa84bee35e651e4c5f09073d668c7aed4cf7e23c3cda84b23"
+dependencies = [
+ "base64 0.22.1",
+ "byteorder",
+ "bytes",
+ "fallible-iterator",
+ "hmac",
+ "md-5",
+ "memchr",
+ "rand 0.8.5",
+ "sha2",
+ "stringprep",
+]
+
 [[package]]
 name = "postgres-protocol2"
 version = "0.1.0"
@@ -4482,7 +4518,18 @@ source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#511f
 dependencies = [
 "bytes",
 "fallible-iterator",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
+]
+
+[[package]]
+name = "postgres-types"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f66ea23a2d0e5734297357705193335e0a957696f34bed2f2faefacb2fec336f"
+dependencies = [
+ "bytes",
+ "fallible-iterator",
+ "postgres-protocol 0.6.7",
 ]

 [[package]]
@@ -4505,9 +4552,9 @@ dependencies = [
 "rustls 0.23.18",
 "rustls-pemfile 2.1.1",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-postgres-rustls",
 "tokio-rustls 0.26.0",
 "tokio-util",
@@ -4522,7 +4569,7 @@ dependencies = [
 "itertools 0.10.5",
 "once_cell",
 "postgres",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "url",
 ]

@@ -4543,7 +4590,7 @@ dependencies = [
 "pprof",
 "regex",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "tracing",
 "utils",
 ]
@@ -4554,7 +4601,7 @@ version = "0.1.0"
 dependencies = [
 "anyhow",
 "camino",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "workspace_hack",
 ]
@@ -4587,7 +4634,7 @@ dependencies = [
 "smallvec",
 "symbolic-demangle",
 "tempfile",
- "thiserror",
+ "thiserror 1.0.69",
 ]

 [[package]]
@@ -4609,15 +4656,6 @@ version = "0.2.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"

-[[package]]
-name = "pq-sys"
-version = "0.6.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f6cc05d7ea95200187117196eee9edd0644424911821aeb28a18ce60ea0b8793"
-dependencies = [
- "vcpkg",
-]
-
 [[package]]
 name = "pq_proto"
 version = "0.1.0"
@@ -4625,10 +4663,10 @@ dependencies = [
 "byteorder",
 "bytes",
 "itertools 0.10.5",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
 "rand 0.8.5",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 ]

@@ -4699,7 +4737,7 @@ dependencies = [
 "memchr",
 "parking_lot 0.12.1",
 "procfs",
- "thiserror",
+ "thiserror 1.0.69",
 ]

 [[package]]
@@ -4869,11 +4907,11 @@ dependencies = [
 "strum",
 "strum_macros",
 "subtle",
- "thiserror",
+ "thiserror 1.0.69",
 "tikv-jemalloc-ctl",
 "tikv-jemallocator",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-postgres2",
 "tokio-rustls 0.26.0",
 "tokio-tungstenite 0.21.0",
@@ -4930,17 +4968,6 @@ dependencies = [
 "proc-macro2",
 ]

-[[package]]
-name = "r2d2"
-version = "0.8.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "51de85fb3fb6524929c8a2eb85e6b6d363de4e8c48f9e2c2eac4944abc181c93"
-dependencies = [
- "log",
- "parking_lot 0.12.1",
- "scheduled-thread-pool",
-]
-
 [[package]]
 name = "rand"
 version = "0.7.3"
@@ -5277,7 +5304,7 @@ dependencies = [
 "http 1.1.0",
 "reqwest",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "tower-service",
 ]

@@ -5297,7 +5324,7 @@ dependencies = [
 "reqwest",
 "reqwest-middleware",
 "retry-policies",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tracing",
 "wasm-timer",
@@ -5313,7 +5340,7 @@ dependencies = [
 "async-trait",
 "getrandom 0.2.11",
 "http 1.1.0",
- "matchit 0.8.4",
+ "matchit",
 "opentelemetry",
 "reqwest",
 "reqwest-middleware",
@@ -5672,7 +5699,7 @@ dependencies = [
 "pageserver_api",
 "parking_lot 0.12.1",
 "postgres",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
 "postgres_backend",
 "postgres_ffi",
 "pprof",
@@ -5692,11 +5719,11 @@ dependencies = [
 "storage_broker",
 "strum",
 "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
 "tikv-jemallocator",
 "tokio",
 "tokio-io-timeout",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-stream",
 "tokio-tar",
 "tokio-util",
@@ -5731,7 +5758,7 @@ dependencies = [
 "reqwest",
 "safekeeper_api",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 "utils",
 "workspace_hack",
 ]
@@ -5755,12 +5782,12 @@ dependencies = [
 ]

 [[package]]
-name = "scheduled-thread-pool"
-version = "0.2.7"
+name = "scoped-futures"
+version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3cbc66816425a074528352f5789333ecff06ca41b36b0b0efdfbb29edc391a19"
+checksum = "1b24aae2d0636530f359e9d5ef0c04669d11c5e756699b27a6a6d845d8329091"
 dependencies = [
- "parking_lot 0.12.1",
+ "pin-project-lite",
 ]

 [[package]]
@@ -5940,7 +5967,7 @@ dependencies = [
 "rand 0.8.5",
 "serde",
 "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
 "time",
 "url",
 "uuid",
@@ -6012,7 +6039,7 @@ checksum = "c7715380eec75f029a4ef7de39a9200e0a63823176b759d055b613f5a87df6a6"
 dependencies = [
 "percent-encoding",
 "serde",
- "thiserror",
+ "thiserror 1.0.69",
 ]

 [[package]]
@@ -6174,7 +6201,7 @@ checksum = "adc4e5204eb1910f40f9cfa375f6f05b68c3abac4b6fd879c8ff5e7ae8a0a085"
 dependencies = [
 "num-bigint",
 "num-traits",
- "thiserror",
+ "thiserror 1.0.69",
 "time",
 ]

@@ -6295,6 +6322,7 @@ dependencies = [
 "clap",
 "control_plane",
 "diesel",
+ "diesel-async",
 "diesel_migrations",
 "fail",
 "futures",
@@ -6309,16 +6337,16 @@ dependencies = [
 "pageserver_api",
 "pageserver_client",
 "postgres_connection",
- "r2d2",
 "rand 0.8.5",
 "reqwest",
 "routerify",
+ "scoped-futures",
 "scopeguard",
 "serde",
 "serde_json",
 "strum",
 "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-util",
 "tracing",
@@ -6365,7 +6393,7 @@ dependencies = [
 "serde_json",
 "storage_controller_client",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-postgres-rustls",
 "tokio-stream",
 "tokio-util",
@@ -6610,7 +6638,16 @@ version = "1.0.69"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b6aaf5339b578ea85b50e080feb250a3e8ae8cfcdff9a461c9ec2904bc923f52"
 dependencies = [
- "thiserror-impl",
+ "thiserror-impl 1.0.69",
+]
+
+[[package]]
+name = "thiserror"
+version = "2.0.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d452f284b73e6d76dd36758a0c8684b1d5be31f92b89d07fd5822175732206fc"
+dependencies = [
+ "thiserror-impl 2.0.11",
 ]

 [[package]]
@@ -6624,6 +6661,17 @@ dependencies = [
 "syn 2.0.90",
 ]

+[[package]]
+name = "thiserror-impl"
+version = "2.0.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "26afc1baea8a989337eeb52b6e72a039780ce45c3edfcc9c5b9d112feeb173c2"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.90",
+]
+
 [[package]]
 name = "thread_local"
 version = "1.1.7"
@@ -6774,13 +6822,13 @@ dependencies = [
 [[package]]
 name = "tokio-epoll-uring"
 version = "0.1.0"
-source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#33e00106a268644d02ba0461bbd64476073b0ee1"
+source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#781989bb540a1408b0b93daa1e9d1fa452195497"
 dependencies = [
 "futures",
 "nix 0.26.4",
 "once_cell",
 "scopeguard",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-util",
 "tracing",
@@ -6824,13 +6872,39 @@ dependencies = [
 "percent-encoding",
 "phf",
 "pin-project-lite",
- "postgres-protocol",
- "postgres-types",
+ "postgres-protocol 0.6.4",
+ "postgres-types 0.2.4",
 "socket2",
 "tokio",
 "tokio-util",
 ]

+[[package]]
+name = "tokio-postgres"
+version = "0.7.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3b5d3742945bc7d7f210693b0c58ae542c6fd47b17adbbda0885f3dcb34a6bdb"
+dependencies = [
+ "async-trait",
+ "byteorder",
+ "bytes",
+ "fallible-iterator",
+ "futures-channel",
+ "futures-util",
+ "log",
+ "parking_lot 0.12.1",
+ "percent-encoding",
+ "phf",
+ "pin-project-lite",
+ "postgres-protocol 0.6.7",
+ "postgres-types 0.2.8",
+ "rand 0.8.5",
+ "socket2",
+ "tokio",
+ "tokio-util",
+ "whoami",
+]
+
 [[package]]
 name = "tokio-postgres-rustls"
 version = "0.12.0"
@@ -6840,7 +6914,7 @@ dependencies = [
 "ring",
 "rustls 0.23.18",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-rustls 0.26.0",
 "x509-certificate",
 ]
@@ -6937,14 +7011,14 @@ dependencies = [

 [[package]]
 name = "tokio-tungstenite"
-version = "0.24.0"
+version = "0.26.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "edc5f74e248dc973e0dbb7b74c7e0d6fcc301c694ff50049504004ef4d0cdcd9"
+checksum = "be4bf6fecd69fcdede0ec680aaf474cdab988f9de6bc73d3758f0160e3b7025a"
 dependencies = [
 "futures-util",
 "log",
 "tokio",
- "tungstenite 0.24.0",
+ "tungstenite 0.26.1",
 ]

 [[package]]
@@ -7004,12 +7078,9 @@ version = "0.12.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "877c5b330756d856ffcc4553ab34a5684481ade925ecc54bcd1bf02b1d0d4d52"
 dependencies = [
- "async-stream",
 "async-trait",
- "axum",
 "base64 0.22.1",
 "bytes",
- "h2 0.4.4",
 "http 1.1.0",
 "http-body 1.0.0",
 "http-body-util",
@@ -7021,7 +7092,6 @@ dependencies = [
 "prost",
 "rustls-native-certs 0.8.0",
 "rustls-pemfile 2.1.1",
- "socket2",
 "tokio",
 "tokio-rustls 0.26.0",
 "tokio-stream",
@@ -7258,16 +7328,16 @@ dependencies = [
 "log",
 "rand 0.8.5",
 "sha1",
- "thiserror",
+ "thiserror 1.0.69",
 "url",
 "utf-8",
 ]

 [[package]]
 name = "tungstenite"
-version = "0.24.0"
+version = "0.26.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "18e5b8366ee7a95b16d32197d0b2604b43a0be89dc5fac9f8e96ccafbaedda8a"
+checksum = "413083a99c579593656008130e29255e54dcaae495be556cc26888f211648c24"
 dependencies = [
 "byteorder",
 "bytes",
@@ -7277,7 +7347,7 @@ dependencies = [
 "log",
 "rand 0.8.5",
 "sha1",
- "thiserror",
+ "thiserror 2.0.11",
 "utf-8",
 ]

@@ -7373,7 +7443,7 @@ dependencies = [
 [[package]]
 name = "uring-common"
 version = "0.1.0"
-source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#33e00106a268644d02ba0461bbd64476073b0ee1"
+source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#781989bb540a1408b0b93daa1e9d1fa452195497"
 dependencies = [
 "bytes",
 "io-uring",
@@ -7472,7 +7542,7 @@ dependencies = [
 "signal-hook",
 "strum",
 "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
 "tokio",
 "tokio-stream",
 "tokio-tar",
@@ -7502,12 +7572,6 @@ version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "830b7e5d4d90034032940e4ace0d9a9a057e7a45cd94e6c007832e39edb82f6d"

-[[package]]
-name = "vcpkg"
-version = "0.2.15"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
-
 [[package]]
 name = "version_check"
 version = "0.9.4"
@@ -7527,7 +7591,7 @@ dependencies = [
 "serde_json",
 "sysinfo",
 "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 "tokio-util",
 "tracing",
 "tracing-subscriber",
@@ -7578,11 +7642,10 @@ dependencies = [
 "remote_storage",
 "serde",
 "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
 "tikv-jemallocator",
 "tokio",
 "tokio-util",
- "tonic",
 "tonic-build",
 "tracing",
 "utils",
@@ -7991,8 +8054,6 @@ version = "0.1.0"
 dependencies = [
 "ahash",
 "anyhow",
- "axum",
- "axum-core",
 "base64 0.13.1",
 "base64 0.21.1",
 "base64ct",
@@ -8073,7 +8134,6 @@ dependencies = [
 "toml_edit",
 "tonic",
 "tower 0.4.13",
- "tower 0.5.2",
 "tracing",
 "tracing-core",
 "url",
@@ -8111,7 +8171,7 @@ dependencies = [
 "ring",
 "signature 2.2.0",
 "spki 0.7.3",
- "thiserror",
+ "thiserror 1.0.69",
 "zeroize",
 ]

@@ -8128,7 +8188,7 @@ dependencies = [
 "nom",
 "oid-registry",
 "rusticata-macros",
- "thiserror",
+ "thiserror 1.0.69",
 "time",
 ]

--- a/Cargo.toml
+++ b/Cargo.toml
@@ -65,7 +65,7 @@ aws-smithy-types = "1.2"
 aws-credential-types = "1.2.0"
 aws-sigv4 = { version = "1.2", features = ["sign-http"] }
 aws-types = "1.3"
-axum = { version = "0.7.9", features = ["ws"] }
+axum = { version = "0.8.1", features = ["ws"] }
 base64 = "0.13.0"
 bincode = "1.3"
 bindgen = "0.70"
@@ -187,7 +187,7 @@ tokio-tar = "0.3"
 tokio-util = { version = "0.7.10", features = ["io", "rt"] }
 toml = "0.8"
 toml_edit = "0.22"
-tonic = {version = "0.12.3", features = ["tls", "tls-roots"]}
+tonic = {version = "0.12.3", default-features = false, features = ["channel", "tls", "tls-roots"]}
 tower = { version = "0.5.2", default-features = false }
 tower-http = { version = "0.6.2", features = ["request-id", "trace"] }
 tower-service = "0.3.3"
--- a/2
+++ b/2
@@ -45,7 +45,7 @@ COPY --chown=nonroot . .

 ARG ADDITIONAL_RUSTFLAGS
 RUN set -e \
-    && PQ_LIB_DIR=$(pwd)/pg_install/v${STABLE_PG_VERSION}/lib RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
+    && RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
      --bin pg_sni_router  \
      --bin pageserver  \
      --bin pagectl  \
--- a/2
+++ b/2
@@ -64,8 +64,6 @@ CARGO_BUILD_FLAGS += $(filter -j1,$(MAKEFLAGS))
 CARGO_CMD_PREFIX += $(if $(filter n,$(MAKEFLAGS)),,+)
 # Force cargo not to print progress bar
 CARGO_CMD_PREFIX += CARGO_TERM_PROGRESS_WHEN=never CI=1
-# Set PQ_LIB_DIR to make sure `storage_controller` get linked with bundled libpq (through diesel)
-CARGO_CMD_PREFIX += PQ_LIB_DIR=$(POSTGRES_INSTALL_DIR)/v16/lib

 CACHEDIR_TAG_CONTENTS := "Signature: 8a477f597d28d172789f06886806bc55"

--- a/README.md
+++ b/README.md
@@ -21,8 +21,10 @@ The Neon storage engine consists of two major components:

 See developer documentation in [SUMMARY.md](/docs/SUMMARY.md) for more information.

-## Running local installation
+## Running a local development environment

+Neon can be run on a workstation for small experiments and to test code changes, by
+following these instructions.

 #### Installing dependencies on Linux
 1. Install build dependencies and other applicable packages
@@ -238,7 +240,7 @@ postgres=# select * from t;
 > cargo neon stop
 ```

-More advanced usages can be found at [Control Plane and Neon Local](./control_plane/README.md).
+More advanced usages can be found at [Local Development Control Plane (`neon_local`))](./control_plane/README.md).

 #### Handling build failures

--- a/compute/compute-node.Dockerfile
+++ b/compute/compute-node.Dockerfile
@@ -67,6 +67,9 @@ RUN cd postgres && \
    # Enable some of contrib extensions
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/autoinc.control && \
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/dblink.control && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgres_fdw.control && \
+    file=/usr/local/pgsql/share/extension/postgres_fdw--1.0.sql && [ -e $file ] && \
+    echo 'GRANT USAGE ON FOREIGN DATA WRAPPER postgres_fdw TO neon_superuser;' >> $file && \
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/bloom.control && \
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/earthdistance.control && \
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/insert_username.control && \
@@ -360,6 +363,8 @@ COPY compute/patches/pgvector.patch /pgvector.patch
 RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.8.0.tar.gz -O pgvector.tar.gz && \
    echo "867a2c328d4928a5a9d6f052cd3bc78c7d60228a9b914ad32aa3db88e9de27b0 pgvector.tar.gz" | sha256sum --check && \
    mkdir pgvector-src && cd pgvector-src && tar xzf ../pgvector.tar.gz --strip-components=1 -C . && \
+    wget https://github.com/pgvector/pgvector/raw/refs/tags/v0.7.4/sql/vector.sql -O ./sql/vector--0.7.4.sql && \
+    echo "10218d05dc02299562252a9484775178b14a1d8edb92a2d1672ef488530f7778 ./sql/vector--0.7.4.sql" | sha256sum --check && \
    patch -p1 < /pgvector.patch && \
    make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" && \
    make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" install && \
@@ -1261,11 +1266,12 @@ RUN set -e \

 #########################################################################################
 #
-# Layers "postgres-exporter" and "sql-exporter"
+# Layers "postgres-exporter", "pgbouncer-exporter", and "sql-exporter"
 #
 #########################################################################################

 FROM quay.io/prometheuscommunity/postgres-exporter:v0.16.0 AS postgres-exporter
+FROM quay.io/prometheuscommunity/pgbouncer-exporter:v0.10.2 AS pgbouncer-exporter

 # Keep the version the same as in build-tools.Dockerfile and
 # test_runner/regress/test_compute_metrics.py.
@@ -1347,9 +1353,6 @@ COPY --from=pg-roaringbitmap-pg-build /pg_roaringbitmap.tar.gz /ext-src
 COPY --from=pg-semver-pg-build /pg_semver.tar.gz /ext-src
 #COPY --from=pg-embedding-pg-build /home/nonroot/pg_embedding-src/ /ext-src
 #COPY --from=wal2json-pg-build /wal2json_2_5.tar.gz /ext-src
-#pg_anon is not supported yet for pg v17 so, don't fail if nothing found
-COPY --from=pg-anon-pg-build /pg_anon.tar.g? /ext-src
-COPY compute/patches/pg_anon.patch /ext-src
 COPY --from=pg-ivm-build /pg_ivm.tar.gz /ext-src
 COPY --from=pg-partman-build /pg_partman.tar.gz /ext-src
 RUN cd /ext-src/ && for f in *.tar.gz; \
@@ -1360,9 +1363,6 @@ RUN cd /ext-src/rum-src && patch -p1 <../rum.patch
 RUN cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
 RUN cd /ext-src/pg_hint_plan-src && patch -p1 < /ext-src/pg_hint_plan_${PG_VERSION}.patch
 COPY --chmod=755 docker-compose/run-tests.sh /run-tests.sh
-RUN case "${PG_VERSION}" in "v17") \
-    echo "postgresql_anonymizer does not yet support PG17" && exit 0;; \
-    esac && patch -p1 </ext-src/pg_anon.patch
 RUN patch -p1 </ext-src/pg_cron.patch
 ENV PATH=/usr/local/pgsql/bin:$PATH
 ENV PGHOST=compute
@@ -1403,6 +1403,7 @@ RUN mkdir -p /etc/local_proxy && chown postgres:postgres /etc/local_proxy

 # Metrics exporter binaries and  configuration files
 COPY --from=postgres-exporter /bin/postgres_exporter /bin/postgres_exporter
+COPY --from=pgbouncer-exporter /bin/pgbouncer_exporter /bin/pgbouncer_exporter
 COPY --from=sql-exporter      /bin/sql_exporter      /bin/sql_exporter

 COPY --chown=postgres compute/etc/postgres_exporter.yml /etc/postgres_exporter.yml
--- a/compute/etc/pgbouncer.ini
+++ b/compute/etc/pgbouncer.ini
@@ -19,6 +19,8 @@ max_prepared_statements=0
 admin_users=postgres
 unix_socket_dir=/tmp/
 unix_socket_mode=0777
+; required for pgbouncer_exporter
+ignore_startup_parameters=extra_float_digits

 ;; Disable connection logging. It produces a lot of logs that no one looks at,
 ;; and we can get similar log entries from the proxy too. We had incidents in
--- a/compute/patches/pgvector.patch
+++ b/compute/patches/pgvector.patch
@@ -1,8 +1,24 @@
+diff --git a/Makefile b/Makefile
+index 7a4b88c..56678af 100644
+--- a/Makefile
+++ b/Makefile
+@@ -3,7 +3,10 @@ EXTVERSION = 0.8.0
+ 
+ MODULE_big = vector
+ DATA = $(wildcard sql/*--*--*.sql)
+-DATA_built = sql/$(EXTENSION)--$(EXTVERSION).sql
+# This change is needed to install different per-version SQL files
+# like pgvector--0.8.0.sql and pgvector--0.7.4.sql
+# The corresponding file is downloaded during the Docker image build process
+DATA_built = sql/$(EXTENSION)--$(EXTVERSION).sql sql/vector--0.7.4.sql
+ OBJS = src/bitutils.o src/bitvec.o src/halfutils.o src/halfvec.o src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/sparsevec.o src/vector.o
+ HEADERS = src/halfvec.h src/sparsevec.h src/vector.h
+ 
 diff --git a/src/hnswbuild.c b/src/hnswbuild.c
-index dcfb2bd..d5189ee 100644
+index b667478..fc1897c 100644
 --- a/src/hnswbuild.c
 +++ b/src/hnswbuild.c
-@@ -860,9 +860,17 @@ HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc)
+@@ -843,9 +843,17 @@ HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc)
 
 	hnswarea = shm_toc_lookup(toc, PARALLEL_KEY_HNSW_AREA, false);
 
@@ -20,7 +36,7 @@ index dcfb2bd..d5189ee 100644
 	/* Close relations within worker */
 	index_close(indexRel, indexLockmode);
 	table_close(heapRel, heapLockmode);
-@@ -1117,12 +1125,38 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
+@@ -1100,12 +1108,38 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
 	SeedRandom(42);
 #endif
 
--- a/compute/vm-image-spec-bookworm.yaml
+++ b/compute/vm-image-spec-bookworm.yaml
@@ -27,6 +27,10 @@ commands:
    user: nobody
    sysvInitAction: respawn
    shell: 'DATA_SOURCE_NAME="user=cloud_admin sslmode=disable dbname=postgres application_name=postgres-exporter" /bin/postgres_exporter --config.file=/etc/postgres_exporter.yml'
+  - name: pgbouncer-exporter
+    user: postgres
+    sysvInitAction: respawn
+    shell: '/bin/pgbouncer_exporter --pgBouncer.connectionString="postgres:///pgbouncer?host=/tmp&port=6432&dbname=pgbouncer&user=pgbouncer"'
  - name: sql-exporter
    user: nobody
    sysvInitAction: respawn
--- a/compute/vm-image-spec-bullseye.yaml
+++ b/compute/vm-image-spec-bullseye.yaml
@@ -27,6 +27,10 @@ commands:
    user: nobody
    sysvInitAction: respawn
    shell: 'DATA_SOURCE_NAME="user=cloud_admin sslmode=disable dbname=postgres application_name=postgres-exporter" /bin/postgres_exporter --config.file=/etc/postgres_exporter.yml'
+  - name: pgbouncer-exporter
+    user: postgres
+    sysvInitAction: respawn
+    shell: '/bin/pgbouncer_exporter --pgBouncer.connectionString="postgres:///pgbouncer?host=/tmp&port=6432&dbname=pgbouncer&user=pgbouncer"'
  - name: sql-exporter
    user: nobody
    sysvInitAction: respawn
--- a/compute_tools/src/bin/fast_import.rs
+++ b/compute_tools/src/bin/fast_import.rs
@@ -58,6 +58,8 @@ struct Args {
    pg_bin_dir: Utf8PathBuf,
    #[clap(long)]
    pg_lib_dir: Utf8PathBuf,
+    #[clap(long)]
+    pg_port: Option<u16>, // port to run postgres on, 5432 is default
 }

 #[serde_with::serde_as]
@@ -74,6 +76,13 @@ enum EncryptionSecret {
    KMS { key_id: String },
 }

+// copied from pageserver_api::config::defaults::DEFAULT_LOCALE to avoid dependency just for a constant
+const DEFAULT_LOCALE: &str = if cfg!(target_os = "macos") {
+    "C"
+} else {
+    "C.UTF-8"
+};
+
 #[tokio::main]
 pub(crate) async fn main() -> anyhow::Result<()> {
    utils::logging::init(
@@ -97,6 +106,10 @@ pub(crate) async fn main() -> anyhow::Result<()> {
    let working_directory = args.working_directory;
    let pg_bin_dir = args.pg_bin_dir;
    let pg_lib_dir = args.pg_lib_dir;
+    let pg_port = args.pg_port.unwrap_or_else(|| {
+        info!("pg_port not specified, using default 5432");
+        5432
+    });

    // Initialize AWS clients only if s3_prefix is specified
    let (aws_config, kms_client) = if args.s3_prefix.is_some() {
@@ -180,7 +193,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
    let superuser = "cloud_admin"; // XXX: this shouldn't be hard-coded
    postgres_initdb::do_run_initdb(postgres_initdb::RunInitdbArgs {
        superuser,
-        locale: "en_US.UTF-8", // XXX: this shouldn't be hard-coded,
+        locale: DEFAULT_LOCALE, // XXX: this shouldn't be hard-coded,
        pg_version,
        initdb_bin: pg_bin_dir.join("initdb").as_ref(),
        library_search_path: &pg_lib_dir, // TODO: is this right? Prob works in compute image, not sure about neon_local.
@@ -197,6 +210,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
    let mut postgres_proc = tokio::process::Command::new(pgbin)
        .arg("-D")
        .arg(&pgdata_dir)
+        .args(["-p", &format!("{pg_port}")])
        .args(["-c", "wal_level=minimal"])
        .args(["-c", "shared_buffers=10GB"])
        .args(["-c", "max_wal_senders=0"])
@@ -216,6 +230,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
            ),
        ])
        .env_clear()
+        .env("LD_LIBRARY_PATH", &pg_lib_dir)
        .stdout(std::process::Stdio::piped())
        .stderr(std::process::Stdio::piped())
        .spawn()
@@ -232,7 +247,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {

    // Create neondb database in the running postgres
    let restore_pg_connstring =
-        format!("host=localhost port=5432 user={superuser} dbname=postgres");
+        format!("host=localhost port={pg_port} user={superuser} dbname=postgres");

    let start_time = std::time::Instant::now();

@@ -314,6 +329,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
            .arg(&source_connection_string)
            // how we run it
            .env_clear()
+            .env("LD_LIBRARY_PATH", &pg_lib_dir)
            .kill_on_drop(true)
            .stdout(std::process::Stdio::piped())
            .stderr(std::process::Stdio::piped())
@@ -347,6 +363,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
            .arg(&dumpdir)
            // how we run it
            .env_clear()
+            .env("LD_LIBRARY_PATH", &pg_lib_dir)
            .kill_on_drop(true)
            .stdout(std::process::Stdio::piped())
            .stderr(std::process::Stdio::piped())
--- a/compute_tools/src/compute.rs
+++ b/compute_tools/src/compute.rs
@@ -41,14 +41,14 @@ use crate::local_proxy;
 use crate::pg_helpers::*;
 use crate::spec::*;
 use crate::spec_apply::ApplySpecPhase::{
-    CreateAndAlterDatabases, CreateAndAlterRoles, CreateAvailabilityCheck, CreateSuperUser,
-    DropInvalidDatabases, DropRoles, HandleNeonExtension, HandleOtherExtensions,
-    RenameAndDeleteDatabases, RenameRoles, RunInEachDatabase,
+    CreateAndAlterDatabases, CreateAndAlterRoles, CreateAvailabilityCheck, CreateSchemaNeon,
+    CreateSuperUser, DropInvalidDatabases, DropRoles, FinalizeDropLogicalSubscriptions,
+    HandleNeonExtension, HandleOtherExtensions, RenameAndDeleteDatabases, RenameRoles,
+    RunInEachDatabase,
 };
 use crate::spec_apply::PerDatabasePhase;
 use crate::spec_apply::PerDatabasePhase::{
-    ChangeSchemaPerms, DeleteDBRoleReferences, DropSubscriptionsForDeletedDatabases,
-    HandleAnonExtension,
+    ChangeSchemaPerms, DeleteDBRoleReferences, DropLogicalSubscriptions, HandleAnonExtension,
 };
 use crate::spec_apply::{apply_operations, MutableApplyContext, DB};
 use crate::sync_sk::{check_if_synced, ping_safekeeper};
@@ -340,6 +340,15 @@ impl ComputeNode {
        self.state.lock().unwrap().status
    }

+    pub fn get_timeline_id(&self) -> Option<TimelineId> {
+        self.state
+            .lock()
+            .unwrap()
+            .pspec
+            .as_ref()
+            .map(|s| s.timeline_id)
+    }
+
    // Remove `pgdata` directory and create it again with right permissions.
    fn create_pgdata(&self) -> Result<()> {
        // Ignore removal error, likely it is a 'No such file or directory (os error 2)'.
@@ -929,6 +938,48 @@ impl ComputeNode {
                .map(|role| (role.name.clone(), role))
                .collect::<HashMap<String, Role>>();

+            // Check if we need to drop subscriptions before starting the endpoint.
+            //
+            // It is important to do this operation exactly once when endpoint starts on a new branch.
+            // Otherwise, we may drop not inherited, but newly created subscriptions.
+            //
+            // We cannot rely only on spec.drop_subscriptions_before_start flag,
+            // because if for some reason compute restarts inside VM,
+            // it will start again with the same spec and flag value.
+            //
+            // To handle this, we save the fact of the operation in the database
+            // in the neon.drop_subscriptions_done table.
+            // If the table does not exist, we assume that the operation was never performed, so we must do it.
+            // If table exists, we check if the operation was performed on the current timelilne.
+            //
+            let mut drop_subscriptions_done = false;
+
+            if spec.drop_subscriptions_before_start {
+                let timeline_id = self.get_timeline_id().context("timeline_id must be set")?;
+                let query = format!("select 1 from neon.drop_subscriptions_done where timeline_id = '{}'", timeline_id);
+
+                info!("Checking if drop subscription operation was already performed for timeline_id: {}", timeline_id);
+
+                drop_subscriptions_done =  match
+                    client.simple_query(&query).await {
+                    Ok(result) => {
+                        matches!(&result[0], postgres::SimpleQueryMessage::Row(_))
+                    },
+                    Err(e) =>
+                    {
+                        match e.code() {
+                            Some(&SqlState::UNDEFINED_TABLE) => false,
+                            _ => {
+                                // We don't expect any other error here, except for the schema/table not existing
+                                error!("Error checking if drop subscription operation was already performed: {}", e);
+                                return Err(e.into());
+                            }
+                        }
+                    }
+                }
+            };
+
+
            let jwks_roles = Arc::new(
                spec.as_ref()
                    .local_proxy_config
@@ -996,7 +1047,7 @@ impl ComputeNode {
                        jwks_roles.clone(),
                        concurrency_token.clone(),
                        db,
-                        [DropSubscriptionsForDeletedDatabases].to_vec(),
+                        [DropLogicalSubscriptions].to_vec(),
                    );

                    Ok(spawn(fut))
@@ -1024,6 +1075,7 @@ impl ComputeNode {
                CreateAndAlterRoles,
                RenameAndDeleteDatabases,
                CreateAndAlterDatabases,
+                CreateSchemaNeon,
            ] {
                info!("Applying phase {:?}", &phase);
                apply_operations(
@@ -1064,6 +1116,17 @@ impl ComputeNode {
                    }

                    let conf = Arc::new(conf);
+                    let mut phases = vec![
+                        DeleteDBRoleReferences,
+                        ChangeSchemaPerms,
+                        HandleAnonExtension,
+                    ];
+
+                    if spec.drop_subscriptions_before_start && !drop_subscriptions_done {
+                        info!("Adding DropLogicalSubscriptions phase because drop_subscriptions_before_start is set");
+                        phases.push(DropLogicalSubscriptions);
+                    }
+
                    let fut = Self::apply_spec_sql_db(
                        spec.clone(),
                        conf,
@@ -1071,12 +1134,7 @@ impl ComputeNode {
                        jwks_roles.clone(),
                        concurrency_token.clone(),
                        db,
-                        [
-                            DeleteDBRoleReferences,
-                            ChangeSchemaPerms,
-                            HandleAnonExtension,
-                        ]
-                        .to_vec(),
+                        phases,
                    );

                    Ok(spawn(fut))
@@ -1088,12 +1146,20 @@ impl ComputeNode {
                handle.await??;
            }

-            for phase in vec![
+            let mut phases = vec![
                HandleOtherExtensions,
-                HandleNeonExtension,
+                HandleNeonExtension, // This step depends on CreateSchemaNeon
                CreateAvailabilityCheck,
                DropRoles,
-            ] {
+            ];
+
+            // This step depends on CreateSchemaNeon
+            if spec.drop_subscriptions_before_start && !drop_subscriptions_done {
+                info!("Adding FinalizeDropLogicalSubscriptions phase because drop_subscriptions_before_start is set");
+                phases.push(FinalizeDropLogicalSubscriptions);
+            }
+
+            for phase in phases {
                debug!("Applying phase {:?}", &phase);
                apply_operations(
                    spec.clone(),
@@ -1463,6 +1529,14 @@ impl ComputeNode {
                        Ok(())
                    },
                )?;
+
+                let postgresql_conf_path = pgdata_path.join("postgresql.conf");
+                if config::line_in_file(
+                    &postgresql_conf_path,
+                    "neon.disable_logical_replication_subscribers=false",
+                )? {
+                    info!("updated postgresql.conf to set neon.disable_logical_replication_subscribers=false");
+                }
                self.pg_reload_conf()?;
            }
            self.post_apply_config()?;
--- a/compute_tools/src/config.rs
+++ b/compute_tools/src/config.rs
@@ -129,6 +129,13 @@ pub fn write_postgres_conf(

    writeln!(file, "neon.extension_server_port={}", extension_server_port)?;

+    if spec.drop_subscriptions_before_start {
+        writeln!(file, "neon.disable_logical_replication_subscribers=true")?;
+    } else {
+        // be explicit about the default value
+        writeln!(file, "neon.disable_logical_replication_subscribers=false")?;
+    }
+
    // This is essential to keep this line at the end of the file,
    // because it is intended to override any settings above.
    writeln!(file, "include_if_exists = 'compute_ctl_temp_override.conf'")?;
--- a/compute_tools/src/http/extract/json.rs
+++ b/compute_tools/src/http/extract/json.rs
@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};

-use axum::{
-    async_trait,
-    extract::{rejection::JsonRejection, FromRequest, Request},
-};
+use axum::extract::{rejection::JsonRejection, FromRequest, Request};
 use compute_api::responses::GenericAPIError;
 use http::StatusCode;

@@ -12,7 +9,6 @@ use http::StatusCode;
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Json<T>(pub T);

-#[async_trait]
 impl<S, T> FromRequest<S> for Json<T>
 where
    axum::Json<T>: FromRequest<S, Rejection = JsonRejection>,
--- a/compute_tools/src/http/extract/path.rs
+++ b/compute_tools/src/http/extract/path.rs
@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};

-use axum::{
-    async_trait,
-    extract::{rejection::PathRejection, FromRequestParts},
-};
+use axum::extract::{rejection::PathRejection, FromRequestParts};
 use compute_api::responses::GenericAPIError;
 use http::{request::Parts, StatusCode};

@@ -12,7 +9,6 @@ use http::{request::Parts, StatusCode};
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Path<T>(pub T);

-#[async_trait]
 impl<S, T> FromRequestParts<S> for Path<T>
 where
    axum::extract::Path<T>: FromRequestParts<S, Rejection = PathRejection>,
--- a/compute_tools/src/http/extract/query.rs
+++ b/compute_tools/src/http/extract/query.rs
@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};

-use axum::{
-    async_trait,
-    extract::{rejection::QueryRejection, FromRequestParts},
-};
+use axum::extract::{rejection::QueryRejection, FromRequestParts};
 use compute_api::responses::GenericAPIError;
 use http::{request::Parts, StatusCode};

@@ -12,7 +9,6 @@ use http::{request::Parts, StatusCode};
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Query<T>(pub T);

-#[async_trait]
 impl<S, T> FromRequestParts<S> for Query<T>
 where
    axum::extract::Query<T>: FromRequestParts<S, Rejection = QueryRejection>,
--- a/compute_tools/src/http/server.rs
+++ b/compute_tools/src/http/server.rs
@@ -55,7 +55,7 @@ async fn serve(port: u16, compute: Arc<ComputeNode>) {
        .route("/database_schema", get(database_schema::get_schema_dump))
        .route("/dbs_and_roles", get(dbs_and_roles::get_catalog_objects))
        .route(
-            "/extension_server/*filename",
+            "/extension_server/{*filename}",
            post(extension_server::download_extension),
        )
        .route("/extensions", post(extensions::install_extension))
--- a/compute_tools/src/spec_apply.rs
+++ b/compute_tools/src/spec_apply.rs
@@ -47,7 +47,7 @@ pub enum PerDatabasePhase {
    DeleteDBRoleReferences,
    ChangeSchemaPerms,
    HandleAnonExtension,
-    DropSubscriptionsForDeletedDatabases,
+    DropLogicalSubscriptions,
 }

 #[derive(Clone, Debug)]
@@ -58,11 +58,13 @@ pub enum ApplySpecPhase {
    CreateAndAlterRoles,
    RenameAndDeleteDatabases,
    CreateAndAlterDatabases,
+    CreateSchemaNeon,
    RunInEachDatabase { db: DB, subphase: PerDatabasePhase },
    HandleOtherExtensions,
    HandleNeonExtension,
    CreateAvailabilityCheck,
    DropRoles,
+    FinalizeDropLogicalSubscriptions,
 }

 pub struct Operation {
@@ -331,7 +333,7 @@ async fn get_operations<'a>(
                            // NB: there could be other db states, which prevent us from dropping
                            // the database. For example, if db is used by any active subscription
                            // or replication slot.
-                            // Such cases are handled in the DropSubscriptionsForDeletedDatabases
+                            // Such cases are handled in the DropLogicalSubscriptions
                            // phase. We do all the cleanup before actually dropping the database.
                            let drop_db_query: String = format!(
                                "DROP DATABASE IF EXISTS {} WITH (FORCE)",
@@ -442,13 +444,19 @@ async fn get_operations<'a>(

            Ok(Box::new(operations))
        }
+        ApplySpecPhase::CreateSchemaNeon => Ok(Box::new(once(Operation {
+            query: String::from("CREATE SCHEMA IF NOT EXISTS neon"),
+            comment: Some(String::from(
+                "create schema for neon extension and utils tables",
+            )),
+        }))),
        ApplySpecPhase::RunInEachDatabase { db, subphase } => {
            match subphase {
-                PerDatabasePhase::DropSubscriptionsForDeletedDatabases => {
+                PerDatabasePhase::DropLogicalSubscriptions => {
                    match &db {
                        DB::UserDB(db) => {
                            let drop_subscription_query: String = format!(
-                                include_str!("sql/drop_subscription_for_drop_dbs.sql"),
+                                include_str!("sql/drop_subscriptions.sql"),
                                datname_str = escape_literal(&db.name),
                            );

@@ -666,10 +674,6 @@ async fn get_operations<'a>(
        }
        ApplySpecPhase::HandleNeonExtension => {
            let operations = vec![
-                Operation {
-                    query: String::from("CREATE SCHEMA IF NOT EXISTS neon"),
-                    comment: Some(String::from("init: add schema for extension")),
-                },
                Operation {
                    query: String::from("CREATE EXTENSION IF NOT EXISTS neon WITH SCHEMA neon"),
                    comment: Some(String::from(
@@ -712,5 +716,9 @@ async fn get_operations<'a>(

            Ok(Box::new(operations))
        }
+        ApplySpecPhase::FinalizeDropLogicalSubscriptions => Ok(Box::new(once(Operation {
+            query: String::from(include_str!("sql/finalize_drop_subscriptions.sql")),
+            comment: None,
+        }))),
    }
 }
--- a/compute_tools/src/sql/drop_subscription_for_drop_dbs.sql
+++ b/compute_tools/src/sql/drop_subscription_for_drop_dbs.sql
--- a/compute_tools/src/sql/finalize_drop_subscriptions.sql
+++ b/compute_tools/src/sql/finalize_drop_subscriptions.sql
@@ -0,0 +1,21 @@
+DO $$
+BEGIN
+    IF NOT EXISTS(
+        SELECT 1
+        FROM pg_catalog.pg_tables
+        WHERE tablename = 'drop_subscriptions_done'
+        AND schemaname = 'neon'
+    )
+    THEN
+        CREATE TABLE neon.drop_subscriptions_done
+        (id serial primary key, timeline_id text);
+    END IF;
+
+    -- preserve the timeline_id of the last drop_subscriptions run
+    -- to ensure that the cleanup of a timeline is executed only once.
+    -- use upsert to avoid the table bloat in case of cascade branching (branch of a branch)
+    INSERT INTO neon.drop_subscriptions_done VALUES (1, current_setting('neon.timeline_id'))
+    ON CONFLICT (id) DO UPDATE
+    SET timeline_id = current_setting('neon.timeline_id');
+END
+$$
--- a/control_plane/README.md
+++ b/control_plane/README.md
@@ -1,6 +1,10 @@
-# Control Plane and Neon Local
+# Local Development Control Plane (`neon_local`)

-This crate contains tools to start a Neon development environment locally. This utility can be used with the `cargo neon` command.
+This crate contains tools to start a Neon development environment locally. This utility can be used with the `cargo neon` command.  This is a convenience to invoke
+the `neon_local` binary.
+
+**Note**: this is a dev/test tool -- a minimal control plane suitable for testing
+code changes locally, but not suitable for running production systems.

 ## Example: Start with Postgres 16

--- a/control_plane/src/bin/neon_local.rs
+++ b/control_plane/src/bin/neon_local.rs
@@ -1357,6 +1357,7 @@ async fn handle_endpoint(subcmd: &EndpointCmd, env: &local_env::LocalEnv) -> Res
                args.pg_version,
                mode,
                !args.update_catalog,
+                false,
            )?;
        }
        EndpointCmd::Start(args) => {
--- a/control_plane/src/endpoint.rs
+++ b/control_plane/src/endpoint.rs
@@ -76,6 +76,7 @@ pub struct EndpointConf {
    http_port: u16,
    pg_version: u32,
    skip_pg_catalog_updates: bool,
+    drop_subscriptions_before_start: bool,
    features: Vec<ComputeFeature>,
 }

@@ -143,6 +144,7 @@ impl ComputeControlPlane {
        pg_version: u32,
        mode: ComputeMode,
        skip_pg_catalog_updates: bool,
+        drop_subscriptions_before_start: bool,
    ) -> Result<Arc<Endpoint>> {
        let pg_port = pg_port.unwrap_or_else(|| self.get_port());
        let http_port = http_port.unwrap_or_else(|| self.get_port() + 1);
@@ -162,6 +164,7 @@ impl ComputeControlPlane {
            // with this we basically test a case of waking up an idle compute, where
            // we also skip catalog updates in the cloud.
            skip_pg_catalog_updates,
+            drop_subscriptions_before_start,
            features: vec![],
        });

@@ -177,6 +180,7 @@ impl ComputeControlPlane {
                pg_port,
                pg_version,
                skip_pg_catalog_updates,
+                drop_subscriptions_before_start,
                features: vec![],
            })?,
        )?;
@@ -240,6 +244,7 @@ pub struct Endpoint {
    // Optimizations
    skip_pg_catalog_updates: bool,

+    drop_subscriptions_before_start: bool,
    // Feature flags
    features: Vec<ComputeFeature>,
 }
@@ -291,6 +296,7 @@ impl Endpoint {
            tenant_id: conf.tenant_id,
            pg_version: conf.pg_version,
            skip_pg_catalog_updates: conf.skip_pg_catalog_updates,
+            drop_subscriptions_before_start: conf.drop_subscriptions_before_start,
            features: conf.features,
        })
    }
@@ -625,6 +631,7 @@ impl Endpoint {
            shard_stripe_size: Some(shard_stripe_size),
            local_proxy_config: None,
            reconfigure_concurrency: 1,
+            drop_subscriptions_before_start: self.drop_subscriptions_before_start,
        };
        let spec_path = self.endpoint_path().join("spec.json");
        std::fs::write(spec_path, serde_json::to_string_pretty(&spec)?)?;
--- a/control_plane/src/pageserver.rs
+++ b/control_plane/src/pageserver.rs
@@ -352,6 +352,21 @@ impl PageServerNode {
                .map(serde_json::from_str)
                .transpose()
                .context("Failed to parse 'compaction_algorithm' json")?,
+            l0_flush_delay_threshold: settings
+                .remove("l0_flush_delay_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_delay_threshold' as an integer")?,
+            l0_flush_wait_upload: settings
+                .remove("l0_flush_wait_upload")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_wait_upload' as a boolean")?,
+            l0_flush_stall_threshold: settings
+                .remove("l0_flush_stall_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_stall_threshold' as an integer")?,
            gc_horizon: settings
                .remove("gc_horizon")
                .map(|x| x.parse::<u64>())
@@ -418,6 +433,26 @@ impl PageServerNode {
                .map(serde_json::from_str)
                .transpose()
                .context("parse `wal_receiver_protocol_override` from json")?,
+            rel_size_v2_enabled: settings
+                .remove("rel_size_v2_enabled")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'rel_size_v2_enabled' as bool")?,
+            gc_compaction_enabled: settings
+                .remove("gc_compaction_enabled")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_enabled' as bool")?,
+            gc_compaction_initial_threshold_kb: settings
+                .remove("gc_compaction_initial_threshold_kb")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_initial_threshold_kb' as integer")?,
+            gc_compaction_ratio_percent: settings
+                .remove("gc_compaction_ratio_percent")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_ratio_percent' as integer")?,
        };
        if !settings.is_empty() {
            bail!("Unrecognized tenant settings: {settings:?}")
--- a/deny.toml
+++ b/deny.toml
@@ -41,8 +41,8 @@ allow = [
    "MIT",
    "MPL-2.0",
    "OpenSSL",
-    "Unicode-DFS-2016",
    "Unicode-3.0",
+    "Zlib",
 ]
 confidence-threshold = 0.8
 exceptions = [
--- a/docker-compose/compute_wrapper/Dockerfile
+++ b/docker-compose/compute_wrapper/Dockerfile
@@ -10,10 +10,7 @@ USER root
 RUN apt-get update &&       \
    apt-get install -y curl \
                       jq   \
-                       python3-pip \
                       netcat-openbsd
-#Faker is required for the pg_anon test
-RUN case $COMPUTE_IMAGE in compute-node-v17) OPT="--break-system-packages";; *) OPT= ;; esac && pip3 install $OPT Faker
 #This is required for the pg_hintplan test
 RUN mkdir -p /ext-src/pg_hint_plan-src && chown postgres /ext-src/pg_hint_plan-src 

--- a/docker-compose/docker-compose.yml
+++ b/docker-compose/docker-compose.yml
@@ -150,8 +150,8 @@ services:
        - REPOSITORY=${REPOSITORY:-neondatabase}
        - COMPUTE_IMAGE=compute-node-v${PG_VERSION:-16}
        - TAG=${TAG:-latest}
-        - http_proxy=$http_proxy
-        - https_proxy=$https_proxy
+        - http_proxy=${http_proxy:-}
+        - https_proxy=${https_proxy:-}
    environment:
      - PG_VERSION=${PG_VERSION:-16}
      #- RUST_BACKTRACE=1
@@ -185,6 +185,8 @@ services:
  neon-test-extensions:
    profiles: ["test-extensions"]
    image: ${REPOSITORY:-neondatabase}/neon-test-extensions-v${PG_TEST_VERSION:-16}:${TAG:-latest}
+    environment:
+      - PGPASSWORD=cloud_admin
    entrypoint:
      - "/bin/bash"
      - "-c"
--- a/docker-compose/docker_compose_test.sh
+++ b/docker-compose/docker_compose_test.sh
@@ -18,14 +18,10 @@ cd $(dirname $0)
 COMPUTE_CONTAINER_NAME=docker-compose-compute-1
 TEST_CONTAINER_NAME=docker-compose-neon-test-extensions-1
 PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -d postgres"
-: ${http_proxy:=}
-: ${https_proxy:=}
-export http_proxy https_proxy

 cleanup() {
    echo "show container information"
    docker ps
-    docker compose --profile test-extensions -f $COMPOSE_FILE logs
    echo "stop containers..."
    docker compose --profile test-extensions -f $COMPOSE_FILE down
 }
@@ -35,12 +31,6 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
    echo "clean up containers if exists"
    cleanup
    PG_TEST_VERSION=$((pg_version < 16 ? 16 : pg_version))
-    # The support of pg_anon not yet added to PG17, so we have to add the corresponding option for other PG versions
-    if [ "${pg_version}" -ne 17 ]; then
-      SPEC_PATH="compute_wrapper/var/db/postgres/specs"
-      mv $SPEC_PATH/spec.json $SPEC_PATH/spec.bak
-      jq '.cluster.settings += [{"name": "session_preload_libraries","value": "anon","vartype": "string"}]' "${SPEC_PATH}/spec.bak" > "${SPEC_PATH}/spec.json"
-    fi
    PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --build -d

    echo "wait until the compute is ready. timeout after 60s. "
@@ -50,7 +40,6 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
        cnt=`expr $cnt + 3`
        if [ $cnt -gt 60 ]; then
            echo "timeout before the compute is ready."
-            cleanup
            exit 1
        fi
        if docker compose --profile test-extensions -f $COMPOSE_FILE logs "compute_is_ready" | grep -q "accepting connections"; then
@@ -62,36 +51,19 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
    done

    if [ $pg_version -ge 16 ]; then
-        echo Enabling trust connection
-        docker exec $COMPUTE_CONTAINER_NAME bash -c "sed -i '\$d' /var/db/postgres/compute/pg_hba.conf && echo -e 'host\t all\t all\t all\t trust' >> /var/db/postgres/compute/pg_hba.conf && psql $PSQL_OPTION -c 'select pg_reload_conf()' "
-        echo Adding postgres role
-        docker exec $COMPUTE_CONTAINER_NAME psql $PSQL_OPTION -c "CREATE ROLE postgres SUPERUSER LOGIN"
        # This is required for the pg_hint_plan test, to prevent flaky log message causing the test to fail
        # It cannot be moved to Dockerfile now because the database directory is created after the start of the container
        echo Adding dummy config
        docker exec $COMPUTE_CONTAINER_NAME touch /var/db/postgres/compute/compute_ctl_temp_override.conf
-        # This block is required for the pg_anon extension test.
-        # The test assumes that it is running on the same host with the postgres engine.
-        # In our case it's not true, that's why we are copying files to the compute node
+        # The following block copies the files for the pg_hintplan test to the compute node for the extension test in an isolated docker-compose environment
        TMPDIR=$(mktemp -d)
-        # Add support for pg_anon for pg_v16
-        if [ $pg_version -ne 17 ]; then
-          docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
-          echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
-          docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
-        rm -rf $TMPDIR
-        fi
-        TMPDIR=$(mktemp -d)
-        # The following block does the same for the pg_hintplan test
        docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
        docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
        rm -rf $TMPDIR
        # We are running tests now
-        if docker exec -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
+        if ! docker exec -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
            $TEST_CONTAINER_NAME /run-tests.sh | tee testout.txt
        then
-            cleanup
-        else
            FAILED=$(tail -1 testout.txt)
            for d in $FAILED
            do
@@ -101,13 +73,7 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
                cat $d/regression.out $d/regression.diffs || true
            done
        rm -rf $FAILED
-        cleanup
        exit 1
        fi
    fi
-    cleanup
-    # Restore the original spec.json
-    if [ "$pg_version" -ne 17 ]; then
-      mv "$SPEC_PATH/spec.bak" "$SPEC_PATH/spec.json"
-    fi
 done
--- a/docker-compose/run-tests.sh
+++ b/docker-compose/run-tests.sh
@@ -7,7 +7,10 @@ LIST=$( (echo -e "${SKIP//","/"\n"}"; ls -d -- *-src) | sort | uniq -u)
 for d in ${LIST}
 do
       [ -d "${d}" ] || continue
-    psql -c "select 1" >/dev/null || break
+       if ! psql -w -c "select 1" >/dev/null; then
+          FAILED="${d} ${FAILED}"
+          break
+       fi
       USE_PGXS=1 make -C "${d}" installcheck || FAILED="${d} ${FAILED}"
 done
 [ -z "${FAILED}" ] && exit 0
--- a/libs/compute_api/src/spec.rs
+++ b/libs/compute_api/src/spec.rs
@@ -138,6 +138,13 @@ pub struct ComputeSpec {
    /// enough spare connections for reconfiguration process to succeed.
    #[serde(default = "default_reconfigure_concurrency")]
    pub reconfigure_concurrency: usize,
+
+    /// If set to true, the compute_ctl will drop all subscriptions before starting the
+    /// compute. This is needed when we start an endpoint on a branch, so that child
+    /// would not compete with parent branch subscriptions
+    /// over the same replication content from publisher.
+    #[serde(default)] // Default false
+    pub drop_subscriptions_before_start: bool,
 }

 /// Feature flag to signal `compute_ctl` to enable certain experimental functionality.
--- a/libs/pageserver_api/src/config.rs
+++ b/libs/pageserver_api/src/config.rs
@@ -120,6 +120,7 @@ pub struct ConfigToml {
    pub no_sync: Option<bool>,
    pub wal_receiver_protocol: PostgresClientProtocol,
    pub page_service_pipelining: PageServicePipeliningConfig,
+    pub get_vectored_concurrent_io: GetVectoredConcurrentIo,
 }

 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -158,6 +159,25 @@ pub enum PageServiceProtocolPipelinedExecutionStrategy {
    Tasks,
 }

+#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+#[serde(tag = "mode", rename_all = "kebab-case")]
+#[serde(deny_unknown_fields)]
+pub enum GetVectoredConcurrentIo {
+    /// The read path is fully sequential: layers are visited
+    /// one after the other and IOs are issued and waited upon
+    /// from the same task that traverses the layers.
+    Sequential,
+    /// The read path still traverses layers sequentially, and
+    /// index blocks will be read into the PS PageCache from
+    /// that task, with waiting.
+    /// But data IOs are dispatched and waited upon from a sidecar
+    /// task so that the traversing task can continue to traverse
+    /// layers while the IOs are in flight.
+    /// If the PS PageCache miss rate is low, this improves
+    /// throughput dramatically.
+    SidecarTask,
+}
+
 pub mod statvfs {
    pub mod mock {
        #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -234,9 +254,21 @@ pub struct TenantConfigToml {
    // Duration::ZERO means automatic compaction is disabled.
    #[serde(with = "humantime_serde")]
    pub compaction_period: Duration,
-    // Level0 delta layer threshold for compaction.
+    /// Level0 delta layer threshold for compaction.
    pub compaction_threshold: usize,
    pub compaction_algorithm: crate::models::CompactionAlgorithmSettings,
+    /// Level0 delta layer threshold at which to delay layer flushes for compaction backpressure,
+    /// such that they take 2x as long, and start waiting for layer flushes during ephemeral layer
+    /// rolls. This helps compaction keep up with WAL ingestion, and avoids read amplification
+    /// blowing up. Should be >compaction_threshold. 0 to disable. Disabled by default.
+    pub l0_flush_delay_threshold: Option<usize>,
+    /// Level0 delta layer threshold at which to stall layer flushes. Must be >compaction_threshold
+    /// to avoid deadlock. 0 to disable. Disabled by default.
+    pub l0_flush_stall_threshold: Option<usize>,
+    /// If true, Level0 delta layer flushes will wait for S3 upload before flushing the next
+    /// layer. This is a temporary backpressure mechanism which should be removed once
+    /// l0_flush_{delay,stall}_threshold is fully enabled.
+    pub l0_flush_wait_upload: bool,
    // Determines how much history is retained, to allow
    // branching and read replicas at an older point in time.
    // The unit is #of bytes of WAL.
@@ -301,6 +333,20 @@ pub struct TenantConfigToml {
    pub timeline_offloading: bool,

    pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
+
+    /// Enable rel_size_v2 for this tenant. Once enabled, the tenant will persist this information into
+    /// `index_part.json`, and it cannot be reversed.
+    pub rel_size_v2_enabled: Option<bool>,
+
+    // gc-compaction related configs
+    /// Enable automatic gc-compaction trigger on this tenant.
+    pub gc_compaction_enabled: bool,
+    /// The initial threshold for gc-compaction in KB. Once the total size of layers below the gc-horizon is above this threshold,
+    /// gc-compaction will be triggered.
+    pub gc_compaction_initial_threshold_kb: u64,
+    /// The ratio that triggers the auto gc-compaction. If (the total size of layers between L2 LSN and gc-horizon) / (size below the L2 LSN)
+    /// is above this ratio, gc-compaction will be triggered.
+    pub gc_compaction_ratio_percent: u64,
 }

 pub mod defaults {
@@ -450,6 +496,11 @@ impl Default for ConfigToml {
                    execution: PageServiceProtocolPipelinedExecutionStrategy::ConcurrentFutures,
                })
            },
+            get_vectored_concurrent_io: if !cfg!(test) {
+                GetVectoredConcurrentIo::Sequential
+            } else {
+                GetVectoredConcurrentIo::SidecarTask
+            },
        }
    }
 }
@@ -475,6 +526,8 @@ pub mod tenant_conf_defaults {
    pub const DEFAULT_COMPACTION_ALGORITHM: crate::models::CompactionAlgorithm =
        crate::models::CompactionAlgorithm::Legacy;

+    pub const DEFAULT_L0_FLUSH_WAIT_UPLOAD: bool = true;
+
    pub const DEFAULT_GC_HORIZON: u64 = 64 * 1024 * 1024;

    // Large DEFAULT_GC_PERIOD is fine as long as PITR_INTERVAL is larger.
@@ -494,6 +547,9 @@ pub mod tenant_conf_defaults {
    // By default ingest enough WAL for two new L0 layers before checking if new image
    // image layers should be created.
    pub const DEFAULT_IMAGE_LAYER_CREATION_CHECK_THRESHOLD: u8 = 2;
+    pub const DEFAULT_GC_COMPACTION_ENABLED: bool = false;
+    pub const DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB: u64 = 10240000;
+    pub const DEFAULT_GC_COMPACTION_RATIO_PERCENT: u64 = 100;
 }

 impl Default for TenantConfigToml {
@@ -510,6 +566,9 @@ impl Default for TenantConfigToml {
            compaction_algorithm: crate::models::CompactionAlgorithmSettings {
                kind: DEFAULT_COMPACTION_ALGORITHM,
            },
+            l0_flush_delay_threshold: None,
+            l0_flush_stall_threshold: None,
+            l0_flush_wait_upload: DEFAULT_L0_FLUSH_WAIT_UPLOAD,
            gc_horizon: DEFAULT_GC_HORIZON,
            gc_period: humantime::parse_duration(DEFAULT_GC_PERIOD)
                .expect("cannot parse default gc period"),
@@ -538,6 +597,10 @@ impl Default for TenantConfigToml {
            lsn_lease_length_for_ts: LsnLease::DEFAULT_LENGTH_FOR_TS,
            timeline_offloading: false,
            wal_receiver_protocol_override: None,
+            rel_size_v2_enabled: None,
+            gc_compaction_enabled: DEFAULT_GC_COMPACTION_ENABLED,
+            gc_compaction_initial_threshold_kb: DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB,
+            gc_compaction_ratio_percent: DEFAULT_GC_COMPACTION_RATIO_PERCENT,
        }
    }
 }
--- a/libs/pageserver_api/src/key.rs
+++ b/libs/pageserver_api/src/key.rs
@@ -464,6 +464,18 @@ pub fn rel_size_to_key(rel: RelTag) -> Key {
    }
 }

+#[inline(always)]
+pub fn rel_size_key_to_rel(key: Key) -> RelTag {
+    assert_eq!(key.field1, 0x00);
+    assert_eq!(key.field6, 0xffff_ffff);
+    RelTag {
+        forknum: key.field5,
+        spcnode: key.field2,
+        dbnode: key.field3,
+        relnode: key.field4,
+    }
+}
+
 impl Key {
    #[inline(always)]
    pub fn is_rel_size_key(&self) -> bool {
@@ -559,6 +571,15 @@ pub fn slru_segment_size_to_key(kind: SlruKind, segno: u32) -> Key {
    }
 }

+#[inline(always)]
+pub fn slru_segment_size_key_to_segno(key: Key) -> u32 {
+    assert_eq!(key.field1, 0x01);
+    assert_eq!(key.field3, 1);
+    assert_eq!(key.field5, 0);
+    assert_eq!(key.field6, 0xffff_ffff);
+    key.field4
+}
+
 impl Key {
    pub fn is_slru_segment_size_key(&self) -> bool {
        self.field1 == 0x01
--- a/libs/pageserver_api/src/models.rs
+++ b/libs/pageserver_api/src/models.rs
@@ -462,6 +462,12 @@ pub struct TenantConfigPatch {
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub compaction_algorithm: FieldPatch<CompactionAlgorithmSettings>,
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_delay_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_stall_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_wait_upload: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub gc_horizon: FieldPatch<u64>,
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub gc_period: FieldPatch<String>,
@@ -497,6 +503,14 @@ pub struct TenantConfigPatch {
    pub timeline_offloading: FieldPatch<bool>,
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub wal_receiver_protocol_override: FieldPatch<PostgresClientProtocol>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub rel_size_v2_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_initial_threshold_kb: FieldPatch<u64>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_ratio_percent: FieldPatch<u64>,
 }

 /// An alternative representation of `pageserver::tenant::TenantConf` with
@@ -510,6 +524,9 @@ pub struct TenantConfig {
    pub compaction_threshold: Option<usize>,
    // defer parsing compaction_algorithm, like eviction_policy
    pub compaction_algorithm: Option<CompactionAlgorithmSettings>,
+    pub l0_flush_delay_threshold: Option<usize>,
+    pub l0_flush_stall_threshold: Option<usize>,
+    pub l0_flush_wait_upload: Option<bool>,
    pub gc_horizon: Option<u64>,
    pub gc_period: Option<String>,
    pub image_creation_threshold: Option<usize>,
@@ -528,6 +545,10 @@ pub struct TenantConfig {
    pub lsn_lease_length_for_ts: Option<String>,
    pub timeline_offloading: Option<bool>,
    pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
+    pub rel_size_v2_enabled: Option<bool>,
+    pub gc_compaction_enabled: Option<bool>,
+    pub gc_compaction_initial_threshold_kb: Option<u64>,
+    pub gc_compaction_ratio_percent: Option<u64>,
 }

 impl TenantConfig {
@@ -539,6 +560,9 @@ impl TenantConfig {
            mut compaction_period,
            mut compaction_threshold,
            mut compaction_algorithm,
+            mut l0_flush_delay_threshold,
+            mut l0_flush_stall_threshold,
+            mut l0_flush_wait_upload,
            mut gc_horizon,
            mut gc_period,
            mut image_creation_threshold,
@@ -557,6 +581,10 @@ impl TenantConfig {
            mut lsn_lease_length_for_ts,
            mut timeline_offloading,
            mut wal_receiver_protocol_override,
+            mut rel_size_v2_enabled,
+            mut gc_compaction_enabled,
+            mut gc_compaction_initial_threshold_kb,
+            mut gc_compaction_ratio_percent,
        } = self;

        patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -567,6 +595,13 @@ impl TenantConfig {
        patch.compaction_period.apply(&mut compaction_period);
        patch.compaction_threshold.apply(&mut compaction_threshold);
        patch.compaction_algorithm.apply(&mut compaction_algorithm);
+        patch
+            .l0_flush_delay_threshold
+            .apply(&mut l0_flush_delay_threshold);
+        patch
+            .l0_flush_stall_threshold
+            .apply(&mut l0_flush_stall_threshold);
+        patch.l0_flush_wait_upload.apply(&mut l0_flush_wait_upload);
        patch.gc_horizon.apply(&mut gc_horizon);
        patch.gc_period.apply(&mut gc_period);
        patch
@@ -601,6 +636,16 @@ impl TenantConfig {
        patch
            .wal_receiver_protocol_override
            .apply(&mut wal_receiver_protocol_override);
+        patch.rel_size_v2_enabled.apply(&mut rel_size_v2_enabled);
+        patch
+            .gc_compaction_enabled
+            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_initial_threshold_kb
+            .apply(&mut gc_compaction_initial_threshold_kb);
+        patch
+            .gc_compaction_ratio_percent
+            .apply(&mut gc_compaction_ratio_percent);

        Self {
            checkpoint_distance,
@@ -609,6 +654,9 @@ impl TenantConfig {
            compaction_period,
            compaction_threshold,
            compaction_algorithm,
+            l0_flush_delay_threshold,
+            l0_flush_stall_threshold,
+            l0_flush_wait_upload,
            gc_horizon,
            gc_period,
            image_creation_threshold,
@@ -627,6 +675,10 @@ impl TenantConfig {
            lsn_lease_length_for_ts,
            timeline_offloading,
            wal_receiver_protocol_override,
+            rel_size_v2_enabled,
+            gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent,
        }
    }
 }
@@ -969,6 +1021,13 @@ pub struct TenantConfigPatchRequest {
    pub config: TenantConfigPatch, // as we have a flattened field, we should reject all unknown fields in it
 }

+#[derive(Serialize, Deserialize, Debug)]
+pub struct TenantWaitLsnRequest {
+    #[serde(flatten)]
+    pub timelines: HashMap<TimelineId, Lsn>,
+    pub timeout: Duration,
+}
+
 /// See [`TenantState::attachment_status`] and the OpenAPI docs for context.
 #[derive(Serialize, Deserialize, Clone)]
 #[serde(tag = "slug", content = "data", rename_all = "snake_case")]
--- a/libs/remote_storage/src/azure_blob.rs
+++ b/libs/remote_storage/src/azure_blob.rs
@@ -377,7 +377,8 @@ impl RemoteStorage for AzureBlobStorage {

                let next_item = next_item?;

-                if timeout_try_cnt >= 2 {
+                // Log a warning if we saw two timeouts in a row before a successful request
+                if timeout_try_cnt > 2 {
                    tracing::warn!("Azure Blob Storage list timed out and succeeded after {} tries", timeout_try_cnt);
                }
                timeout_try_cnt = 1;
--- a/libs/utils/src/env.rs
+++ b/libs/utils/src/env.rs
@@ -2,6 +2,7 @@

 use std::{fmt::Display, str::FromStr};

+/// For types `V` that implement [`FromStr`].
 pub fn var<V, E>(varname: &str) -> Option<V>
 where
    V: FromStr<Err = E>,
@@ -10,7 +11,9 @@ where
    match std::env::var(varname) {
        Ok(s) => Some(
            s.parse()
-                .map_err(|e| format!("failed to parse env var {varname}: {e:#}"))
+                .map_err(|e| {
+                    format!("failed to parse env var {varname} using FromStr::parse: {e:#}")
+                })
                .unwrap(),
        ),
        Err(std::env::VarError::NotPresent) => None,
@@ -19,3 +22,24 @@ where
        }
    }
 }
+
+/// For types `V` that implement [`serde::de::DeserializeOwned`].
+pub fn var_serde_json_string<V>(varname: &str) -> Option<V>
+where
+    V: serde::de::DeserializeOwned,
+{
+    match std::env::var(varname) {
+        Ok(s) => Some({
+            let value = serde_json::Value::String(s);
+            serde_json::from_value(value)
+                .map_err(|e| {
+                    format!("failed to parse env var {varname} as a serde_json json string: {e:#}")
+                })
+                .unwrap()
+        }),
+        Err(std::env::VarError::NotPresent) => None,
+        Err(std::env::VarError::NotUnicode(_)) => {
+            panic!("env var {varname} is not unicode")
+        }
+    }
+}
--- a/libs/utils/src/failpoint_support.rs
+++ b/libs/utils/src/failpoint_support.rs
@@ -11,31 +11,55 @@ use tracing::*;

 /// Declare a failpoint that can use to `pause` failpoint action.
 /// We don't want to block the executor thread, hence, spawn_blocking + await.
+///
+/// Optionally pass a cancellation token, and this failpoint will drop out of
+/// its pause when the cancellation token fires. This is useful for testing
+/// cases where we would like to block something, but test its clean shutdown behavior.
+/// The macro evaluates to a Result in that case, where Ok(()) is the case
+/// where the failpoint was not paused, and Err() is the case where cancellation
+/// token fired while evaluating the failpoint.
+///
+/// Remember to unpause the failpoint in the test; until that happens, one of the
+/// limited number of spawn_blocking thread pool threads is leaked.
 #[macro_export]
 macro_rules! pausable_failpoint {
-    ($name:literal) => {
+    ($name:literal) => {{
        if cfg!(feature = "testing") {
-            tokio::task::spawn_blocking({
-                let current = tracing::Span::current();
+            let cancel = ::tokio_util::sync::CancellationToken::new();
+            let _ = $crate::pausable_failpoint!($name, &cancel);
+        }
+    }};
+    ($name:literal, $cancel:expr) => {{
+        if cfg!(feature = "testing") {
+            let failpoint_fut = ::tokio::task::spawn_blocking({
+                let current = ::tracing::Span::current();
                move || {
                    let _entered = current.entered();
-                    tracing::info!("at failpoint {}", $name);
-                    fail::fail_point!($name);
+                    ::tracing::info!("at failpoint {}", $name);
+                    ::fail::fail_point!($name);
+                }
+            });
+            let cancel_fut = async move {
+                $cancel.cancelled().await;
+            };
+            ::tokio::select! {
+                res = failpoint_fut => {
+                    res.expect("spawn_blocking");
+                    // continue with execution
+                    Ok(())
+                },
+                _ = cancel_fut => {
+                    Err(())
                }
-            })
-            .await
-            .expect("spawn_blocking");
-        }
-    };
-    ($name:literal, $cond:expr) => {
-        if cfg!(feature = "testing") {
-            if $cond {
-                pausable_failpoint!($name)
            }
+        } else {
+            Ok(())
        }
-    };
+    }};
 }

+pub use pausable_failpoint;
+
 /// use with fail::cfg("$name", "return(2000)")
 ///
 /// The effect is similar to a "sleep(2000)" action, i.e. we sleep for the
--- a/libs/utils/src/sync/gate.rs
+++ b/libs/utils/src/sync/gate.rs
@@ -64,6 +64,12 @@ pub struct GateGuard {
    gate: Arc<GateInner>,
 }

+impl GateGuard {
+    pub fn try_clone(&self) -> Result<Self, GateError> {
+        Gate::enter_impl(self.gate.clone())
+    }
+}
+
 impl Drop for GateGuard {
    fn drop(&mut self) {
        if self.gate.closing.load(Ordering::Relaxed) {
@@ -107,11 +113,11 @@ impl Gate {
    /// to avoid blocking close() indefinitely: typically types that contain a Gate will
    /// also contain a CancellationToken.
    pub fn enter(&self) -> Result<GateGuard, GateError> {
-        let permit = self
-            .inner
-            .sem
-            .try_acquire()
-            .map_err(|_| GateError::GateClosed)?;
+        Self::enter_impl(self.inner.clone())
+    }
+
+    fn enter_impl(gate: Arc<GateInner>) -> Result<GateGuard, GateError> {
+        let permit = gate.sem.try_acquire().map_err(|_| GateError::GateClosed)?;

        // we now have the permit, let's disable the normal raii functionality and leave
        // "returning" the permit to our GateGuard::drop.
@@ -122,7 +128,7 @@ impl Gate {

        Ok(GateGuard {
            span_at_enter: tracing::Span::current(),
-            gate: self.inner.clone(),
+            gate,
        })
    }

@@ -252,4 +258,39 @@ mod tests {
        // Attempting to enter() is still forbidden
        gate.enter().expect_err("enter should fail finishing close");
    }
+
+    #[tokio::test(start_paused = true)]
+    async fn clone_gate_guard() {
+        let gate = Gate::default();
+        let forever = Duration::from_secs(24 * 7 * 365);
+
+        let guard1 = gate.enter().expect("gate isn't closed");
+
+        let guard2 = guard1.try_clone().expect("gate isn't clsoed");
+
+        let mut close_fut = std::pin::pin!(gate.close());
+
+        tokio::time::timeout(forever, &mut close_fut)
+            .await
+            .unwrap_err();
+
+        // we polled close_fut once, that should prevent all later enters and clones
+        gate.enter().unwrap_err();
+        guard1.try_clone().unwrap_err();
+        guard2.try_clone().unwrap_err();
+
+        // guard2 keeps gate open even if guard1 is closed
+        drop(guard1);
+        tokio::time::timeout(forever, &mut close_fut)
+            .await
+            .unwrap_err();
+
+        drop(guard2);
+
+        // now that the last guard is dropped, closing should complete
+        close_fut.await;
+
+        // entering is still forbidden
+        gate.enter().expect_err("enter should stilll fail");
+    }
 }
--- a/libs/vm_monitor/src/dispatcher.rs
+++ b/libs/vm_monitor/src/dispatcher.rs
@@ -7,7 +7,7 @@
 //! (notifying it of upscale).

 use anyhow::{bail, Context};
-use axum::extract::ws::{Message, WebSocket};
+use axum::extract::ws::{Message, Utf8Bytes, WebSocket};
 use futures::{
    stream::{SplitSink, SplitStream},
    SinkExt, StreamExt,
@@ -82,21 +82,21 @@ impl Dispatcher {

        let highest_shared_version = match monitor_range.highest_shared_version(&agent_range) {
            Ok(version) => {
-                sink.send(Message::Text(
+                sink.send(Message::Text(Utf8Bytes::from(
                    serde_json::to_string(&ProtocolResponse::Version(version)).unwrap(),
-                ))
+                )))
                .await
                .context("failed to notify agent of negotiated protocol version")?;
                version
            }
            Err(e) => {
-                sink.send(Message::Text(
+                sink.send(Message::Text(Utf8Bytes::from(
                    serde_json::to_string(&ProtocolResponse::Error(format!(
                        "Received protocol version range {} which does not overlap with {}",
                        agent_range, monitor_range
                    )))
                    .unwrap(),
-                ))
+                )))
                .await
                .context("failed to notify agent of no overlap between protocol version ranges")?;
                Err(e).context("error determining suitable protocol version range")?
@@ -126,7 +126,7 @@ impl Dispatcher {

        let json = serde_json::to_string(&message).context("failed to serialize message")?;
        self.sink
-            .send(Message::Text(json))
+            .send(Message::Text(Utf8Bytes::from(json)))
            .await
            .context("stream error sending message")
    }
--- a/libs/wal_decoder/Cargo.toml
+++ b/libs/wal_decoder/Cargo.toml
@@ -17,7 +17,6 @@ postgres_ffi.workspace = true
 serde.workspace = true
 thiserror.workspace = true
 tokio = { workspace = true, features = ["io-util"] }
-tonic.workspace = true
 tracing.workspace = true
 utils.workspace = true
 workspace_hack = { version = "0.1", path = "../../workspace_hack" }
--- a/libs/wal_decoder/src/models.rs
+++ b/libs/wal_decoder/src/models.rs
@@ -45,7 +45,7 @@ pub mod proto {
    #![allow(clippy::derive_partial_eq_without_eq)]
    // The generated ValueMeta has a `len` method generate for its `len` field.
    #![allow(clippy::len_without_is_empty)]
-    tonic::include_proto!("interpreted_wal");
+    include!(concat!(env!("OUT_DIR"), concat!("/interpreted_wal.rs")));
 }

 #[derive(Copy, Clone, Serialize, Deserialize)]
--- a/pageserver/client/src/mgmt_api.rs
+++ b/pageserver/client/src/mgmt_api.rs
@@ -763,4 +763,19 @@ impl Client {
            .await
            .map_err(Error::ReceiveBody)
    }
+
+    pub async fn wait_lsn(
+        &self,
+        tenant_shard_id: TenantShardId,
+        request: TenantWaitLsnRequest,
+    ) -> Result<StatusCode> {
+        let uri = format!(
+            "{}/v1/tenant/{tenant_shard_id}/wait_lsn",
+            self.mgmt_api_endpoint,
+        );
+
+        self.request_noerror(Method::POST, uri, request)
+            .await
+            .map(|resp| resp.status())
+    }
 }
--- a/pageserver/compaction/src/simulator/draw.rs
+++ b/pageserver/compaction/src/simulator/draw.rs
@@ -160,9 +160,12 @@ pub fn draw_history<W: std::io::Write>(history: &[LayerTraceEvent], mut output:

        // Fill in and thicken rectangle if it's an
        // image layer so that we can see it.
-        let mut style = Style::default();
-        style.fill = Fill::Color(rgb(0x80, 0x80, 0x80));
-        style.stroke = Stroke::Color(rgb(0, 0, 0), 0.5);
+        let mut style = Style {
+            fill: Fill::Color(rgb(0x80, 0x80, 0x80)),
+            stroke: Stroke::Color(rgb(0, 0, 0), 0.5),
+            opacity: 1.0,
+            stroke_opacity: 1.0,
+        };

        let y_start = lsn_max - lsn_start;
        let y_end = lsn_max - lsn_end;
@@ -214,10 +217,6 @@ pub fn draw_history<W: std::io::Write>(history: &[LayerTraceEvent], mut output:
        files_seen.insert(f);
    }

-    let mut record_style = Style::default();
-    record_style.fill = Fill::Color(rgb(0x80, 0x80, 0x80));
-    record_style.stroke = Stroke::None;
-
    writeln!(svg, "{}", EndSvg)?;

    let mut layer_events_str = String::new();
--- a/pageserver/pagebench/src/cmd/getpage_latest_lsn.rs
+++ b/pageserver/pagebench/src/cmd/getpage_latest_lsn.rs
@@ -13,7 +13,7 @@ use rand::prelude::*;
 use tokio::task::JoinSet;
 use tracing::info;

-use std::collections::HashSet;
+use std::collections::{HashSet, VecDeque};
 use std::future::Future;
 use std::num::NonZeroUsize;
 use std::pin::Pin;
@@ -63,6 +63,10 @@ pub(crate) struct Args {
    #[clap(long)]
    set_io_mode: Option<pageserver_api::models::virtual_file::IoMode>,

+    /// Queue depth generated in each client.
+    #[clap(long, default_value = "1")]
+    queue_depth: NonZeroUsize,
+
    targets: Option<Vec<TenantTimelineId>>,
 }

@@ -298,6 +302,7 @@ async fn main_impl(
            start_work_barrier.wait().await;
            let client_start = Instant::now();
            let mut ticks_processed = 0;
+            let mut inflight = VecDeque::new();
            while !cancel.is_cancelled() {
                // Detect if a request took longer than the RPS rate
                if let Some(period) = &rps_period {
@@ -311,31 +316,37 @@ async fn main_impl(
                    ticks_processed = periods_passed_until_now;
                }

-                let start = Instant::now();
-                let req = {
-                    let mut rng = rand::thread_rng();
-                    let r = &ranges[weights.sample(&mut rng)];
-                    let key: i128 = rng.gen_range(r.start..r.end);
-                    let key = Key::from_i128(key);
-                    assert!(key.is_rel_block_key());
-                    let (rel_tag, block_no) = key
-                        .to_rel_block()
-                        .expect("we filter non-rel-block keys out above");
-                    PagestreamGetPageRequest {
-                        hdr: PagestreamRequest {
-                            reqid: 0,
-                            request_lsn: if rng.gen_bool(args.req_latest_probability) {
-                                Lsn::MAX
-                            } else {
-                                r.timeline_lsn
+                while inflight.len() < args.queue_depth.get() {
+                    let start = Instant::now();
+                    let req = {
+                        let mut rng = rand::thread_rng();
+                        let r = &ranges[weights.sample(&mut rng)];
+                        let key: i128 = rng.gen_range(r.start..r.end);
+                        let key = Key::from_i128(key);
+                        assert!(key.is_rel_block_key());
+                        let (rel_tag, block_no) = key
+                            .to_rel_block()
+                            .expect("we filter non-rel-block keys out above");
+                        PagestreamGetPageRequest {
+                            hdr: PagestreamRequest {
+                                reqid: 0,
+                                request_lsn: if rng.gen_bool(args.req_latest_probability) {
+                                    Lsn::MAX
+                                } else {
+                                    r.timeline_lsn
+                                },
+                                not_modified_since: r.timeline_lsn,
                            },
-                            not_modified_since: r.timeline_lsn,
-                        },
-                        rel: rel_tag,
-                        blkno: block_no,
-                    }
-                };
-                client.getpage(req).await.unwrap();
+                            rel: rel_tag,
+                            blkno: block_no,
+                        }
+                    };
+                    client.getpage_send(req).await.unwrap();
+                    inflight.push_back(start);
+                }
+
+                let start = inflight.pop_front().unwrap();
+                client.getpage_recv().await.unwrap();
                let end = Instant::now();
                live_stats.request_done();
                ticks_processed += 1;
--- a/pageserver/src/basebackup.rs
+++ b/pageserver/src/basebackup.rs
@@ -25,6 +25,7 @@ use tokio_tar::{Builder, EntryType, Header};

 use crate::context::RequestContext;
 use crate::pgdatadir_mapping::Version;
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::Timeline;
 use pageserver_api::reltag::{RelTag, SlruKind};

@@ -123,6 +124,13 @@ where
        full_backup,
        replica,
        ctx,
+        io_concurrency: IoConcurrency::spawn_from_conf(
+            timeline.conf,
+            timeline
+                .gate
+                .enter()
+                .map_err(|e| BasebackupError::Server(e.into()))?,
+        ),
    };
    basebackup
        .send_tarball()
@@ -144,6 +152,7 @@ where
    full_backup: bool,
    replica: bool,
    ctx: &'a RequestContext,
+    io_concurrency: IoConcurrency,
 }

 /// A sink that accepts SLRU blocks ordered by key and forwards
@@ -303,7 +312,7 @@ where
            for part in slru_partitions.parts {
                let blocks = self
                    .timeline
-                    .get_vectored(part, self.lsn, self.ctx)
+                    .get_vectored(part, self.lsn, self.io_concurrency.clone(), self.ctx)
                    .await
                    .map_err(|e| BasebackupError::Server(e.into()))?;

@@ -358,7 +367,7 @@ where
        let start_time = Instant::now();
        let aux_files = self
            .timeline
-            .list_aux_files(self.lsn, self.ctx)
+            .list_aux_files(self.lsn, self.ctx, self.io_concurrency.clone())
            .await
            .map_err(|e| BasebackupError::Server(e.into()))?;
        let aux_scan_time = start_time.elapsed();
@@ -422,7 +431,7 @@ where
        }
        let repl_origins = self
            .timeline
-            .get_replorigins(self.lsn, self.ctx)
+            .get_replorigins(self.lsn, self.ctx, self.io_concurrency.clone())
            .await
            .map_err(|e| BasebackupError::Server(e.into()))?;
        let n_origins = repl_origins.len();
@@ -489,7 +498,13 @@ where
            for blknum in startblk..endblk {
                let img = self
                    .timeline
-                    .get_rel_page_at_lsn(src, blknum, Version::Lsn(self.lsn), self.ctx)
+                    .get_rel_page_at_lsn(
+                        src,
+                        blknum,
+                        Version::Lsn(self.lsn),
+                        self.ctx,
+                        self.io_concurrency.clone(),
+                    )
                    .await
                    .map_err(|e| BasebackupError::Server(e.into()))?;
                segment_data.extend_from_slice(&img[..]);
--- a/pageserver/src/bin/pageserver.rs
+++ b/pageserver/src/bin/pageserver.rs
@@ -135,6 +135,7 @@ fn main() -> anyhow::Result<()> {
    info!(?conf.virtual_file_io_mode, "starting with virtual_file IO mode");
    info!(?conf.wal_receiver_protocol, "starting with WAL receiver protocol");
    info!(?conf.page_service_pipelining, "starting with page service pipelining config");
+    info!(?conf.get_vectored_concurrent_io, "starting with get_vectored IO concurrency config");

    // The tenants directory contains all the pageserver local disk state.
    // Create if not exists and make sure all the contents are durable before proceeding.
--- a/pageserver/src/config.rs
+++ b/pageserver/src/config.rs
@@ -191,6 +191,8 @@ pub struct PageServerConf {
    pub wal_receiver_protocol: PostgresClientProtocol,

    pub page_service_pipelining: pageserver_api::config::PageServicePipeliningConfig,
+
+    pub get_vectored_concurrent_io: pageserver_api::config::GetVectoredConcurrentIo,
 }

 /// Token for authentication to safekeepers
@@ -352,6 +354,7 @@ impl PageServerConf {
            no_sync,
            wal_receiver_protocol,
            page_service_pipelining,
+            get_vectored_concurrent_io,
        } = config_toml;

        let mut conf = PageServerConf {
@@ -396,6 +399,7 @@ impl PageServerConf {
            import_pgdata_aws_endpoint_url,
            wal_receiver_protocol,
            page_service_pipelining,
+            get_vectored_concurrent_io,

            // ------------------------------------------------------------
            // fields that require additional validation or custom handling
--- a/pageserver/src/http/routes.rs
+++ b/pageserver/src/http/routes.rs
@@ -10,6 +10,7 @@ use std::time::Duration;

 use anyhow::{anyhow, Context, Result};
 use enumset::EnumSet;
+use futures::future::join_all;
 use futures::StreamExt;
 use futures::TryFutureExt;
 use humantime::format_rfc3339;
@@ -40,6 +41,7 @@ use pageserver_api::models::TenantShardSplitRequest;
 use pageserver_api::models::TenantShardSplitResponse;
 use pageserver_api::models::TenantSorting;
 use pageserver_api::models::TenantState;
+use pageserver_api::models::TenantWaitLsnRequest;
 use pageserver_api::models::TimelineArchivalConfigRequest;
 use pageserver_api::models::TimelineCreateRequestMode;
 use pageserver_api::models::TimelineCreateRequestModeImportPgdata;
@@ -84,6 +86,7 @@ use crate::tenant::remote_timeline_client::list_remote_tenant_shards;
 use crate::tenant::remote_timeline_client::list_remote_timelines;
 use crate::tenant::secondary::SecondaryController;
 use crate::tenant::size::ModelInputs;
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::storage_layer::LayerAccessStatsReset;
 use crate::tenant::storage_layer::LayerName;
 use crate::tenant::timeline::import_pgdata;
@@ -94,6 +97,8 @@ use crate::tenant::timeline::CompactOptions;
 use crate::tenant::timeline::CompactRequest;
 use crate::tenant::timeline::CompactionError;
 use crate::tenant::timeline::Timeline;
+use crate::tenant::timeline::WaitLsnTimeout;
+use crate::tenant::timeline::WaitLsnWaiter;
 use crate::tenant::GetTimelineError;
 use crate::tenant::OffloadedTimeline;
 use crate::tenant::{LogicalSizeCalculationCause, PageReconstructError};
@@ -2789,6 +2794,63 @@ async fn secondary_download_handler(
    json_response(status, progress)
 }

+async fn wait_lsn_handler(
+    mut request: Request<Body>,
+    cancel: CancellationToken,
+) -> Result<Response<Body>, ApiError> {
+    let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
+    let wait_lsn_request: TenantWaitLsnRequest = json_request(&mut request).await?;
+
+    let state = get_state(&request);
+    let tenant = state
+        .tenant_manager
+        .get_attached_tenant_shard(tenant_shard_id)?;
+
+    let mut wait_futures = Vec::default();
+    for timeline in tenant.list_timelines() {
+        let Some(lsn) = wait_lsn_request.timelines.get(&timeline.timeline_id) else {
+            continue;
+        };
+
+        let fut = {
+            let timeline = timeline.clone();
+            let ctx = RequestContext::new(TaskKind::MgmtRequest, DownloadBehavior::Error);
+            async move {
+                timeline
+                    .wait_lsn(
+                        *lsn,
+                        WaitLsnWaiter::HttpEndpoint,
+                        WaitLsnTimeout::Custom(wait_lsn_request.timeout),
+                        &ctx,
+                    )
+                    .await
+            }
+        };
+        wait_futures.push(fut);
+    }
+
+    if wait_futures.is_empty() {
+        return json_response(StatusCode::NOT_FOUND, ());
+    }
+
+    let all_done = tokio::select! {
+        results = join_all(wait_futures) => {
+            results.iter().all(|res| res.is_ok())
+        },
+        _ = cancel.cancelled() => {
+            return Err(ApiError::Cancelled);
+        }
+    };
+
+    let status = if all_done {
+        StatusCode::OK
+    } else {
+        StatusCode::ACCEPTED
+    };
+
+    json_response(status, ())
+}
+
 async fn secondary_status_handler(
    request: Request<Body>,
    _cancel: CancellationToken,
@@ -2938,8 +3000,15 @@ async fn list_aux_files(
        active_timeline_of_active_tenant(&state.tenant_manager, tenant_shard_id, timeline_id)
            .await?;

+    let io_concurrency = IoConcurrency::spawn_from_conf(
+        state.conf,
+        timeline.gate.enter().map_err(|_| ApiError::Cancelled)?,
+    );
+
    let ctx = RequestContext::new(TaskKind::MgmtRequest, DownloadBehavior::Download);
-    let files = timeline.list_aux_files(body.lsn, &ctx).await?;
+    let files = timeline
+        .list_aux_files(body.lsn, &ctx, io_concurrency)
+        .await?;
    json_response(StatusCode::OK, files)
 }

@@ -3569,6 +3638,9 @@ pub fn make_router(
        .post("/v1/tenant/:tenant_shard_id/secondary/download", |r| {
            api_handler(r, secondary_download_handler)
        })
+        .post("/v1/tenant/:tenant_shard_id/wait_lsn", |r| {
+            api_handler(r, wait_lsn_handler)
+        })
        .put("/v1/tenant/:tenant_shard_id/break", |r| {
            testing_api_handler("set tenant state to broken", r, handle_tenant_break)
        })
--- a/pageserver/src/metrics.rs
+++ b/pageserver/src/metrics.rs
@@ -38,6 +38,9 @@ pub(crate) enum StorageTimeOperation {
    #[strum(serialize = "layer flush")]
    LayerFlush,

+    #[strum(serialize = "layer flush delay")]
+    LayerFlushDelay,
+
    #[strum(serialize = "compact")]
    Compact,

@@ -126,73 +129,6 @@ pub(crate) static INITDB_RUN_TIME: Lazy<Histogram> = Lazy::new(|| {
    .expect("failed to define metric")
 });

-// Metrics collected on operations on the storage repository.
-#[derive(
-    Clone, Copy, enum_map::Enum, strum_macros::EnumString, strum_macros::Display, IntoStaticStr,
-)]
-pub(crate) enum GetKind {
-    Singular,
-    Vectored,
-}
-
-pub(crate) struct ReconstructTimeMetrics {
-    singular: Histogram,
-    vectored: Histogram,
-}
-
-pub(crate) static RECONSTRUCT_TIME: Lazy<ReconstructTimeMetrics> = Lazy::new(|| {
-    let inner = register_histogram_vec!(
-        "pageserver_getpage_reconstruct_seconds",
-        "Time spent in reconstruct_value (reconstruct a page from deltas)",
-        &["get_kind"],
-        CRITICAL_OP_BUCKETS.into(),
-    )
-    .expect("failed to define a metric");
-
-    ReconstructTimeMetrics {
-        singular: inner.with_label_values(&[GetKind::Singular.into()]),
-        vectored: inner.with_label_values(&[GetKind::Vectored.into()]),
-    }
-});
-
-impl ReconstructTimeMetrics {
-    pub(crate) fn for_get_kind(&self, get_kind: GetKind) -> &Histogram {
-        match get_kind {
-            GetKind::Singular => &self.singular,
-            GetKind::Vectored => &self.vectored,
-        }
-    }
-}
-
-pub(crate) struct ReconstructDataTimeMetrics {
-    singular: Histogram,
-    vectored: Histogram,
-}
-
-impl ReconstructDataTimeMetrics {
-    pub(crate) fn for_get_kind(&self, get_kind: GetKind) -> &Histogram {
-        match get_kind {
-            GetKind::Singular => &self.singular,
-            GetKind::Vectored => &self.vectored,
-        }
-    }
-}
-
-pub(crate) static GET_RECONSTRUCT_DATA_TIME: Lazy<ReconstructDataTimeMetrics> = Lazy::new(|| {
-    let inner = register_histogram_vec!(
-        "pageserver_getpage_get_reconstruct_data_seconds",
-        "Time spent in get_reconstruct_value_data",
-        &["get_kind"],
-        CRITICAL_OP_BUCKETS.into(),
-    )
-    .expect("failed to define a metric");
-
-    ReconstructDataTimeMetrics {
-        singular: inner.with_label_values(&[GetKind::Singular.into()]),
-        vectored: inner.with_label_values(&[GetKind::Vectored.into()]),
-    }
-});
-
 pub(crate) struct GetVectoredLatency {
    map: EnumMap<TaskKind, Option<Histogram>>,
 }
@@ -2550,12 +2486,19 @@ impl StorageTimeMetricsTimer {
        }
    }

-    /// Record the time from creation to now.
-    pub fn stop_and_record(self) {
-        let duration = self.start.elapsed().as_secs_f64();
-        self.metrics.timeline_sum.inc_by(duration);
+    /// Returns the elapsed duration of the timer.
+    pub fn elapsed(&self) -> Duration {
+        self.start.elapsed()
+    }
+
+    /// Record the time from creation to now and return it.
+    pub fn stop_and_record(self) -> Duration {
+        let duration = self.elapsed();
+        let seconds = duration.as_secs_f64();
+        self.metrics.timeline_sum.inc_by(seconds);
        self.metrics.timeline_count.inc();
-        self.metrics.global_histogram.observe(duration);
+        self.metrics.global_histogram.observe(seconds);
+        duration
    }

    /// Turns this timer into a timer, which will always record -- usually this means recording
@@ -2575,6 +2518,13 @@ impl Drop for AlwaysRecordingStorageTimeMetricsTimer {
    }
 }

+impl AlwaysRecordingStorageTimeMetricsTimer {
+    /// Returns the elapsed duration of the timer.
+    pub fn elapsed(&self) -> Duration {
+        self.0.as_ref().expect("not dropped yet").elapsed()
+    }
+}
+
 /// Timing facilities for an globally histogrammed metric, which is supported by per tenant and
 /// timeline total sum and count.
 #[derive(Clone, Debug)]
@@ -2627,6 +2577,7 @@ pub(crate) struct TimelineMetrics {
    shard_id: String,
    timeline_id: String,
    pub flush_time_histo: StorageTimeMetrics,
+    pub flush_delay_histo: StorageTimeMetrics,
    pub flush_wait_upload_time_gauge: Gauge,
    pub compact_time_histo: StorageTimeMetrics,
    pub create_images_time_histo: StorageTimeMetrics,
@@ -2673,6 +2624,12 @@ impl TimelineMetrics {
            &shard_id,
            &timeline_id,
        );
+        let flush_delay_histo = StorageTimeMetrics::new(
+            StorageTimeOperation::LayerFlushDelay,
+            &tenant_id,
+            &shard_id,
+            &timeline_id,
+        );
        let flush_wait_upload_time_gauge = FLUSH_WAIT_UPLOAD_TIME
            .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
            .unwrap();
@@ -2821,6 +2778,7 @@ impl TimelineMetrics {
            shard_id,
            timeline_id,
            flush_time_histo,
+            flush_delay_histo,
            flush_wait_upload_time_gauge,
            compact_time_histo,
            create_images_time_histo,
@@ -3919,7 +3877,6 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
    });

    // Custom
-    Lazy::force(&RECONSTRUCT_TIME);
    Lazy::force(&BASEBACKUP_QUERY_TIME);
    Lazy::force(&COMPUTE_COMMANDS_COUNTERS);
    Lazy::force(&tokio_epoll_uring::THREAD_LOCAL_METRICS_STORAGE);
--- a/pageserver/src/page_service.rs
+++ b/pageserver/src/page_service.rs
@@ -39,6 +39,7 @@ use tokio::io::{AsyncWriteExt, BufWriter};
 use tokio::task::JoinHandle;
 use tokio_util::sync::CancellationToken;
 use tracing::*;
+use utils::sync::gate::{Gate, GateGuard};
 use utils::sync::spsc_fold;
 use utils::{
    auth::{Claims, Scope, SwappableJwtAuth},
@@ -61,6 +62,7 @@ use crate::task_mgr::{self, COMPUTE_REQUEST_RUNTIME};
 use crate::tenant::mgr::ShardSelector;
 use crate::tenant::mgr::TenantManager;
 use crate::tenant::mgr::{GetActiveTenantError, GetTenantError, ShardResolveResult};
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::timeline::{self, WaitLsnError};
 use crate::tenant::GetTimelineError;
 use crate::tenant::PageReconstructError;
@@ -90,6 +92,7 @@ pub struct Listener {
 pub struct Connections {
    cancel: CancellationToken,
    tasks: tokio::task::JoinSet<ConnectionHandlerResult>,
+    gate: Gate,
 }

 pub fn spawn(
@@ -110,6 +113,7 @@ pub fn spawn(
    let task = COMPUTE_REQUEST_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
        "libpq listener",
        libpq_listener_main(
+            conf,
            tenant_manager,
            pg_auth,
            tcp_listener,
@@ -134,11 +138,16 @@ impl Listener {
 }
 impl Connections {
    pub(crate) async fn shutdown(self) {
-        let Self { cancel, mut tasks } = self;
+        let Self {
+            cancel,
+            mut tasks,
+            gate,
+        } = self;
        cancel.cancel();
        while let Some(res) = tasks.join_next().await {
            Self::handle_connection_completion(res);
        }
+        gate.close().await;
    }

    fn handle_connection_completion(res: Result<anyhow::Result<()>, tokio::task::JoinError>) {
@@ -158,7 +167,9 @@ impl Connections {
 /// Returns Ok(()) upon cancellation via `cancel`, returning the set of
 /// open connections.
 ///
+#[allow(clippy::too_many_arguments)]
 pub async fn libpq_listener_main(
+    conf: &'static PageServerConf,
    tenant_manager: Arc<TenantManager>,
    auth: Option<Arc<SwappableJwtAuth>>,
    listener: tokio::net::TcpListener,
@@ -168,9 +179,15 @@ pub async fn libpq_listener_main(
    listener_cancel: CancellationToken,
 ) -> Connections {
    let connections_cancel = CancellationToken::new();
+    let connections_gate = Gate::default();
    let mut connection_handler_tasks = tokio::task::JoinSet::default();

    loop {
+        let gate_guard = match connections_gate.enter() {
+            Ok(guard) => guard,
+            Err(_) => break,
+        };
+
        let accepted = tokio::select! {
            biased;
            _ = listener_cancel.cancelled() => break,
@@ -190,6 +207,7 @@ pub async fn libpq_listener_main(
                let connection_ctx = listener_ctx
                    .detached_child(TaskKind::PageRequestHandler, DownloadBehavior::Download);
                connection_handler_tasks.spawn(page_service_conn_main(
+                    conf,
                    tenant_manager.clone(),
                    local_auth,
                    socket,
@@ -197,6 +215,7 @@ pub async fn libpq_listener_main(
                    pipelining_config.clone(),
                    connection_ctx,
                    connections_cancel.child_token(),
+                    gate_guard,
                ));
            }
            Err(err) => {
@@ -211,13 +230,16 @@ pub async fn libpq_listener_main(
    Connections {
        cancel: connections_cancel,
        tasks: connection_handler_tasks,
+        gate: connections_gate,
    }
 }

 type ConnectionHandlerResult = anyhow::Result<()>;

 #[instrument(skip_all, fields(peer_addr))]
+#[allow(clippy::too_many_arguments)]
 async fn page_service_conn_main(
+    conf: &'static PageServerConf,
    tenant_manager: Arc<TenantManager>,
    auth: Option<Arc<SwappableJwtAuth>>,
    socket: tokio::net::TcpStream,
@@ -225,6 +247,7 @@ async fn page_service_conn_main(
    pipelining_config: PageServicePipeliningConfig,
    connection_ctx: RequestContext,
    cancel: CancellationToken,
+    gate_guard: GateGuard,
 ) -> ConnectionHandlerResult {
    let _guard = LIVE_CONNECTIONS
        .with_label_values(&["page_service"])
@@ -274,11 +297,13 @@ async fn page_service_conn_main(
    // But it's in a shared crate, so, we store connection_ctx inside PageServerHandler
    // and create the per-query context in process_query ourselves.
    let mut conn_handler = PageServerHandler::new(
+        conf,
        tenant_manager,
        auth,
        pipelining_config,
        connection_ctx,
        cancel.clone(),
+        gate_guard,
    );
    let pgbackend = PostgresBackend::new_from_io(socket, peer_addr, auth_type, None)?;

@@ -310,6 +335,7 @@ async fn page_service_conn_main(
 }

 struct PageServerHandler {
+    conf: &'static PageServerConf,
    auth: Option<Arc<SwappableJwtAuth>>,
    claims: Option<Claims>,

@@ -325,6 +351,8 @@ struct PageServerHandler {
    timeline_handles: Option<TimelineHandles>,

    pipelining_config: PageServicePipeliningConfig,
+
+    gate_guard: GateGuard,
 }

 struct TimelineHandles {
@@ -634,19 +662,23 @@ impl BatchedFeMessage {

 impl PageServerHandler {
    pub fn new(
+        conf: &'static PageServerConf,
        tenant_manager: Arc<TenantManager>,
        auth: Option<Arc<SwappableJwtAuth>>,
        pipelining_config: PageServicePipeliningConfig,
        connection_ctx: RequestContext,
        cancel: CancellationToken,
+        gate_guard: GateGuard,
    ) -> Self {
        PageServerHandler {
+            conf,
            auth,
            claims: None,
            connection_ctx,
            timeline_handles: Some(TimelineHandles::new(tenant_manager)),
            cancel,
            pipelining_config,
+            gate_guard,
        }
    }

@@ -1015,6 +1047,7 @@ impl PageServerHandler {
        &mut self,
        pgb_writer: &mut PostgresBackend<IO>,
        batch: BatchedFeMessage,
+        io_concurrency: IoConcurrency,
        cancel: &CancellationToken,
        protocol_version: PagestreamProtocolVersion,
        ctx: &RequestContext,
@@ -1084,6 +1117,7 @@ impl PageServerHandler {
                                &*shard.upgrade()?,
                                effective_request_lsn,
                                pages,
+                                io_concurrency,
                                ctx,
                            )
                            .instrument(span.clone())
@@ -1288,6 +1322,17 @@ impl PageServerHandler {
            }
        }

+        let io_concurrency = IoConcurrency::spawn_from_conf(
+            self.conf,
+            match self.gate_guard.try_clone() {
+                Ok(guard) => guard,
+                Err(_) => {
+                    info!("shutdown request received in page handler");
+                    return Err(QueryError::Shutdown);
+                }
+            },
+        );
+
        let pgb_reader = pgb
            .split()
            .context("implementation error: split pgb into reader and writer")?;
@@ -1309,6 +1354,7 @@ impl PageServerHandler {
                    request_span,
                    pipelining_config,
                    protocol_version,
+                    io_concurrency,
                    &ctx,
                )
                .await
@@ -1322,6 +1368,7 @@ impl PageServerHandler {
                    timeline_handles,
                    request_span,
                    protocol_version,
+                    io_concurrency,
                    &ctx,
                )
                .await
@@ -1349,6 +1396,7 @@ impl PageServerHandler {
        mut timeline_handles: TimelineHandles,
        request_span: Span,
        protocol_version: PagestreamProtocolVersion,
+        io_concurrency: IoConcurrency,
        ctx: &RequestContext,
    ) -> (
        (PostgresBackendReader<IO>, TimelineHandles),
@@ -1383,7 +1431,14 @@ impl PageServerHandler {
            };

            let err = self
-                .pagesteam_handle_batched_message(pgb_writer, msg, &cancel, protocol_version, ctx)
+                .pagesteam_handle_batched_message(
+                    pgb_writer,
+                    msg,
+                    io_concurrency.clone(),
+                    &cancel,
+                    protocol_version,
+                    ctx,
+                )
                .await;
            match err {
                Ok(()) => {}
@@ -1407,6 +1462,7 @@ impl PageServerHandler {
        request_span: Span,
        pipelining_config: PageServicePipeliningConfigPipelined,
        protocol_version: PagestreamProtocolVersion,
+        io_concurrency: IoConcurrency,
        ctx: &RequestContext,
    ) -> (
        (PostgresBackendReader<IO>, TimelineHandles),
@@ -1550,6 +1606,7 @@ impl PageServerHandler {
                    self.pagesteam_handle_batched_message(
                        pgb_writer,
                        batch,
+                        io_concurrency.clone(),
                        &cancel,
                        protocol_version,
                        &ctx,
@@ -1651,6 +1708,7 @@ impl PageServerHandler {
                .wait_lsn(
                    not_modified_since,
                    crate::tenant::timeline::WaitLsnWaiter::PageService,
+                    timeline::WaitLsnTimeout::Default,
                    ctx,
                )
                .await?;
@@ -1806,6 +1864,7 @@ impl PageServerHandler {
        timeline: &Timeline,
        effective_lsn: Lsn,
        requests: smallvec::SmallVec<[BatchedGetPageRequest; 1]>,
+        io_concurrency: IoConcurrency,
        ctx: &RequestContext,
    ) -> Vec<Result<(PagestreamBeMessage, SmgrOpTimer), BatchedPageStreamError>> {
        debug_assert_current_span_has_tenant_and_timeline_id();
@@ -1832,6 +1891,7 @@ impl PageServerHandler {
            .get_rel_page_at_lsn_batched(
                requests.iter().map(|p| (&p.req.rel, &p.req.blkno)),
                effective_lsn,
+                io_concurrency,
                ctx,
            )
            .await;
@@ -1985,6 +2045,7 @@ impl PageServerHandler {
                .wait_lsn(
                    lsn,
                    crate::tenant::timeline::WaitLsnWaiter::PageService,
+                    crate::tenant::timeline::WaitLsnTimeout::Default,
                    ctx,
                )
                .await?;
--- a/pageserver/src/pgdatadir_mapping.rs
+++ b/pageserver/src/pgdatadir_mapping.rs
@@ -17,19 +17,21 @@ use crate::span::{
    debug_assert_current_span_has_tenant_and_timeline_id,
    debug_assert_current_span_has_tenant_and_timeline_id_no_shard_id,
 };
+use crate::tenant::storage_layer::{IoConcurrency, ValuesReconstructState};
 use crate::tenant::timeline::GetVectoredError;
 use anyhow::{ensure, Context};
 use bytes::{Buf, Bytes, BytesMut};
 use enum_map::Enum;
 use itertools::Itertools;
-use pageserver_api::key::Key;
 use pageserver_api::key::{
    dbdir_key_range, rel_block_to_key, rel_dir_to_key, rel_key_range, rel_size_to_key,
    relmap_file_key, repl_origin_key, repl_origin_key_range, slru_block_to_key, slru_dir_to_key,
-    slru_segment_key_range, slru_segment_size_to_key, twophase_file_key, twophase_key_range,
-    CompactKey, AUX_FILES_KEY, CHECKPOINT_KEY, CONTROLFILE_KEY, DBDIR_KEY, TWOPHASEDIR_KEY,
+    slru_segment_key_range, slru_segment_size_key_to_segno, slru_segment_size_to_key,
+    twophase_file_key, twophase_key_range, CompactKey, AUX_FILES_KEY, CHECKPOINT_KEY,
+    CONTROLFILE_KEY, DBDIR_KEY, TWOPHASEDIR_KEY,
 };
-use pageserver_api::keyspace::SparseKeySpace;
+use pageserver_api::key::{rel_size_key_to_rel, Key};
+use pageserver_api::keyspace::{KeySpaceRandomAccum, SparseKeySpace};
 use pageserver_api::record::NeonWalRecord;
 use pageserver_api::reltag::{BlockNumber, RelTag, SlruKind};
 use pageserver_api::shard::ShardIdentity;
@@ -109,10 +111,21 @@ pub(crate) enum CollectKeySpaceError {
    Decode(#[from] DeserializeError),
    #[error(transparent)]
    PageRead(PageReconstructError),
+    #[error(transparent)]
+    GetVectored(GetVectoredError),
    #[error("cancelled")]
    Cancelled,
 }

+impl From<GetVectoredError> for CollectKeySpaceError {
+    fn from(err: GetVectoredError) -> Self {
+        match err {
+            GetVectoredError::Cancelled => Self::Cancelled,
+            err => Self::GetVectored(err),
+        }
+    }
+}
+
 impl From<PageReconstructError> for CollectKeySpaceError {
    fn from(err: PageReconstructError) -> Self {
        match err {
@@ -200,6 +213,7 @@ impl Timeline {
        blknum: BlockNumber,
        version: Version<'_>,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> Result<Bytes, PageReconstructError> {
        match version {
            Version::Lsn(effective_lsn) => {
@@ -208,6 +222,7 @@ impl Timeline {
                    .get_rel_page_at_lsn_batched(
                        pages.iter().map(|(tag, blknum)| (tag, blknum)),
                        effective_lsn,
+                        io_concurrency.clone(),
                        ctx,
                    )
                    .await;
@@ -246,6 +261,7 @@ impl Timeline {
        &self,
        pages: impl ExactSizeIterator<Item = (&RelTag, &BlockNumber)>,
        effective_lsn: Lsn,
+        io_concurrency: IoConcurrency,
        ctx: &RequestContext,
    ) -> Vec<Result<Bytes, PageReconstructError>> {
        debug_assert_current_span_has_tenant_and_timeline_id();
@@ -309,7 +325,10 @@ impl Timeline {
            acc.to_keyspace()
        };

-        match self.get_vectored(keyspace, effective_lsn, ctx).await {
+        match self
+            .get_vectored(keyspace, effective_lsn, io_concurrency, ctx)
+            .await
+        {
            Ok(results) => {
                for (key, res) in results {
                    let mut key_slots = keys_slots.remove(&key).unwrap().into_iter();
@@ -889,9 +908,15 @@ impl Timeline {
        &self,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> Result<HashMap<String, Bytes>, PageReconstructError> {
        let kv = self
-            .scan(KeySpace::single(Key::metadata_aux_key_range()), lsn, ctx)
+            .scan(
+                KeySpace::single(Key::metadata_aux_key_range()),
+                lsn,
+                ctx,
+                io_concurrency,
+            )
            .await?;
        let mut result = HashMap::new();
        let mut sz = 0;
@@ -914,8 +939,9 @@ impl Timeline {
        &self,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> Result<(), PageReconstructError> {
-        self.list_aux_files_v2(lsn, ctx).await?;
+        self.list_aux_files_v2(lsn, ctx, io_concurrency).await?;
        Ok(())
    }

@@ -923,17 +949,24 @@ impl Timeline {
        &self,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> Result<HashMap<String, Bytes>, PageReconstructError> {
-        self.list_aux_files_v2(lsn, ctx).await
+        self.list_aux_files_v2(lsn, ctx, io_concurrency).await
    }

    pub(crate) async fn get_replorigins(
        &self,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> Result<HashMap<RepOriginId, Lsn>, PageReconstructError> {
        let kv = self
-            .scan(KeySpace::single(repl_origin_key_range()), lsn, ctx)
+            .scan(
+                KeySpace::single(repl_origin_key_range()),
+                lsn,
+                ctx,
+                io_concurrency,
+            )
            .await?;
        let mut result = HashMap::new();
        for (k, v) in kv {
@@ -1050,11 +1083,30 @@ impl Timeline {
                .into_iter()
                .collect();
            rels.sort_unstable();
+            let mut relsize_keys_to_collect = KeySpaceRandomAccum::new();
            for rel in rels {
                let relsize_key = rel_size_to_key(rel);
-                let mut buf = self.get(relsize_key, lsn, ctx).await?;
+                relsize_keys_to_collect.add_key(relsize_key);
+            }
+            // Skip the vectored-read max key check by using `get_vectored_impl`.
+            let io_concurrency = IoConcurrency::spawn_from_conf(
+                self.conf,
+                self.gate
+                    .enter()
+                    .map_err(|_| CollectKeySpaceError::Cancelled)?,
+            );
+            let res = self
+                .get_vectored_impl(
+                    relsize_keys_to_collect.consume_keyspace(),
+                    lsn,
+                    &mut ValuesReconstructState::new(io_concurrency),
+                    ctx,
+                )
+                .await?;
+            for (relsize_key, buf) in res {
+                let mut buf = buf?;
                let relsize = buf.get_u32_le();
-
+                let rel = rel_size_key_to_rel(relsize_key);
                result.add_range(rel_block_to_key(rel, 0)..rel_block_to_key(rel, relsize));
                result.add_key(relsize_key);
            }
@@ -1073,11 +1125,30 @@ impl Timeline {
                let dir = SlruSegmentDirectory::des(&buf)?;
                let mut segments: Vec<u32> = dir.segments.iter().cloned().collect();
                segments.sort_unstable();
+                let mut segsize_keys_to_collect = KeySpaceRandomAccum::new();
                for segno in segments {
                    let segsize_key = slru_segment_size_to_key(kind, segno);
-                    let mut buf = self.get(segsize_key, lsn, ctx).await?;
+                    segsize_keys_to_collect.add_key(segsize_key);
+                }
+                // Skip the vectored-read max key check by using `get_vectored_impl`.
+                let io_concurrency = IoConcurrency::spawn_from_conf(
+                    self.conf,
+                    self.gate
+                        .enter()
+                        .map_err(|_| CollectKeySpaceError::Cancelled)?,
+                );
+                let res = self
+                    .get_vectored_impl(
+                        segsize_keys_to_collect.consume_keyspace(),
+                        lsn,
+                        &mut ValuesReconstructState::new(io_concurrency),
+                        ctx,
+                    )
+                    .await?;
+                for (segsize_key, buf) in res {
+                    let mut buf = buf?;
                    let segsize = buf.get_u32_le();
-
+                    let segno = slru_segment_size_key_to_segno(segsize_key);
                    result.add_range(
                        slru_block_to_key(kind, segno, 0)..slru_block_to_key(kind, segno, segsize),
                    );
@@ -2432,7 +2503,11 @@ mod tests {
            ("foo/bar2".to_string(), Bytes::from_static(b"content2")),
        ]);

-        let readback = tline.list_aux_files(Lsn(0x1008), &ctx).await?;
+        let io_concurrency = IoConcurrency::spawn_for_test();
+
+        let readback = tline
+            .list_aux_files(Lsn(0x1008), &ctx, io_concurrency.clone())
+            .await?;
        assert_eq!(readback, expect_1008);

        // Second modification: update one key, remove the other
@@ -2444,11 +2519,15 @@ mod tests {
        let expect_2008 =
            HashMap::from([("foo/bar1".to_string(), Bytes::from_static(b"content3"))]);

-        let readback = tline.list_aux_files(Lsn(0x2008), &ctx).await?;
+        let readback = tline
+            .list_aux_files(Lsn(0x2008), &ctx, io_concurrency.clone())
+            .await?;
        assert_eq!(readback, expect_2008);

        // Reading back in time works
-        let readback = tline.list_aux_files(Lsn(0x1008), &ctx).await?;
+        let readback = tline
+            .list_aux_files(Lsn(0x1008), &ctx, io_concurrency.clone())
+            .await?;
        assert_eq!(readback, expect_1008);

        Ok(())
--- a/pageserver/src/tenant.rs
+++ b/pageserver/src/tenant.rs
@@ -37,6 +37,8 @@ use remote_timeline_client::manifest::{
    OffloadedTimelineManifest, TenantManifest, LATEST_TENANT_MANIFEST_VERSION,
 };
 use remote_timeline_client::UploadQueueNotReadyError;
+use remote_timeline_client::FAILED_REMOTE_OP_RETRIES;
+use remote_timeline_client::FAILED_UPLOAD_WARN_THRESHOLD;
 use std::collections::BTreeMap;
 use std::fmt;
 use std::future::Future;
@@ -2558,7 +2560,12 @@ impl Tenant {
                    // sizes etc. and that would get confused if the previous page versions
                    // are not in the repository yet.
                    ancestor_timeline
-                        .wait_lsn(*lsn, timeline::WaitLsnWaiter::Tenant, ctx)
+                        .wait_lsn(
+                            *lsn,
+                            timeline::WaitLsnWaiter::Tenant,
+                            timeline::WaitLsnTimeout::Default,
+                            ctx,
+                        )
                        .await
                        .map_err(|e| match e {
                            e @ (WaitLsnError::Timeout(_) | WaitLsnError::BadState { .. }) => {
@@ -5308,27 +5315,37 @@ impl Tenant {
            return Ok(());
        }

-        upload_tenant_manifest(
-            &self.remote_storage,
-            &self.tenant_shard_id,
-            self.generation,
-            &manifest,
+        // Remote storage does no retries internally, so wrap it
+        match backoff::retry(
+            || async {
+                upload_tenant_manifest(
+                    &self.remote_storage,
+                    &self.tenant_shard_id,
+                    self.generation,
+                    &manifest,
+                    &self.cancel,
+                )
+                .await
+            },
+            |_e| self.cancel.is_cancelled(),
+            FAILED_UPLOAD_WARN_THRESHOLD,
+            FAILED_REMOTE_OP_RETRIES,
+            "uploading tenant manifest",
            &self.cancel,
        )
        .await
-        .map_err(|e| {
-            if self.cancel.is_cancelled() {
-                TenantManifestError::Cancelled
-            } else {
-                TenantManifestError::RemoteStorage(e)
+        {
+            None => Err(TenantManifestError::Cancelled),
+            Some(Err(_)) if self.cancel.is_cancelled() => Err(TenantManifestError::Cancelled),
+            Some(Err(e)) => Err(TenantManifestError::RemoteStorage(e)),
+            Some(Ok(_)) => {
+                // Store the successfully uploaded manifest, so that future callers can avoid
+                // re-uploading the same thing.
+                *guard = Some(manifest);
+
+                Ok(())
            }
-        })?;
-
-        // Store the successfully uploaded manifest, so that future callers can avoid
-        // re-uploading the same thing.
-        *guard = Some(manifest);
-
-        Ok(())
+        }
    }
 }

@@ -5453,6 +5470,9 @@ pub(crate) mod harness {
                compaction_period: Some(tenant_conf.compaction_period),
                compaction_threshold: Some(tenant_conf.compaction_threshold),
                compaction_algorithm: Some(tenant_conf.compaction_algorithm),
+                l0_flush_delay_threshold: tenant_conf.l0_flush_delay_threshold,
+                l0_flush_stall_threshold: tenant_conf.l0_flush_stall_threshold,
+                l0_flush_wait_upload: Some(tenant_conf.l0_flush_wait_upload),
                gc_horizon: Some(tenant_conf.gc_horizon),
                gc_period: Some(tenant_conf.gc_period),
                image_creation_threshold: Some(tenant_conf.image_creation_threshold),
@@ -5475,6 +5495,12 @@ pub(crate) mod harness {
                lsn_lease_length_for_ts: Some(tenant_conf.lsn_lease_length_for_ts),
                timeline_offloading: Some(tenant_conf.timeline_offloading),
                wal_receiver_protocol_override: tenant_conf.wal_receiver_protocol_override,
+                rel_size_v2_enabled: tenant_conf.rel_size_v2_enabled,
+                gc_compaction_enabled: Some(tenant_conf.gc_compaction_enabled),
+                gc_compaction_initial_threshold_kb: Some(
+                    tenant_conf.gc_compaction_initial_threshold_kb,
+                ),
+                gc_compaction_ratio_percent: Some(tenant_conf.gc_compaction_ratio_percent),
            }
        }
    }
@@ -5708,7 +5734,7 @@ mod tests {
    use pageserver_api::value::Value;
    use pageserver_compaction::helpers::overlaps_with;
    use rand::{thread_rng, Rng};
-    use storage_layer::PersistentLayerKey;
+    use storage_layer::{IoConcurrency, PersistentLayerKey};
    use tests::storage_layer::ValuesReconstructState;
    use tests::timeline::{GetVectoredError, ShutdownMode};
    use timeline::{CompactOptions, DeltaLayerTestDesc};
@@ -6489,6 +6515,7 @@ mod tests {
    async fn test_get_vectored() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_get_vectored").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_test_timeline(TIMELINE_ID, Lsn(0x08), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -6553,7 +6580,7 @@ mod tests {
                .get_vectored_impl(
                    read.clone(),
                    reads_lsn,
-                    &mut ValuesReconstructState::new(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                    &ctx,
                )
                .await;
@@ -6600,6 +6627,7 @@ mod tests {
        let harness = TenantHarness::create("test_get_vectored_aux_files").await?;

        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_empty_timeline(TIMELINE_ID, Lsn(0), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -6634,7 +6662,7 @@ mod tests {
            .get_vectored_impl(
                aux_keyspace.clone(),
                read_lsn,
-                &mut ValuesReconstructState::new(),
+                &mut ValuesReconstructState::new(io_concurrency.clone()),
                &ctx,
            )
            .await;
@@ -6682,6 +6710,7 @@ mod tests {
        )
        .await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let mut current_key = Key::from_hex("010000000033333333444444445500000000").unwrap();
        let gap_at_key = current_key.add(100);
@@ -6782,7 +6811,7 @@ mod tests {
            .get_vectored_impl(
                read.clone(),
                current_lsn,
-                &mut ValuesReconstructState::new(),
+                &mut ValuesReconstructState::new(io_concurrency.clone()),
                &ctx,
            )
            .await?;
@@ -6825,6 +6854,7 @@ mod tests {
    async fn test_get_vectored_ancestor_descent() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_get_vectored_on_lsn_axis").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let start_key = Key::from_hex("010000000033333333444444445500000000").unwrap();
        let end_key = start_key.add(1000);
@@ -6917,7 +6947,7 @@ mod tests {
                        ranges: vec![child_gap_at_key..child_gap_at_key.next()],
                    },
                    query_lsn,
-                    &mut ValuesReconstructState::new(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                    &ctx,
                )
                .await;
@@ -7363,6 +7393,7 @@ mod tests {
    async fn test_metadata_scan() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_metadata_scan").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_test_timeline(TIMELINE_ID, Lsn(0x10), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -7416,7 +7447,7 @@ mod tests {
                .get_vectored_impl(
                    keyspace.clone(),
                    lsn,
-                    &mut ValuesReconstructState::default(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                    &ctx,
                )
                .await?
@@ -7531,6 +7562,7 @@ mod tests {
        let harness = TenantHarness::create("test_aux_file_e2e").await.unwrap();

        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let mut lsn = Lsn(0x08);

@@ -7550,7 +7582,10 @@ mod tests {
        }

        // we can read everything from the storage
-        let files = tline.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = tline
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
        assert_eq!(
            files.get("pg_logical/mappings/test1"),
            Some(&bytes::Bytes::from_static(b"first"))
@@ -7566,7 +7601,10 @@ mod tests {
            modification.commit(&ctx).await.unwrap();
        }

-        let files = tline.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = tline
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
        assert_eq!(
            files.get("pg_logical/mappings/test2"),
            Some(&bytes::Bytes::from_static(b"second"))
@@ -7577,7 +7615,10 @@ mod tests {
            .await
            .unwrap();

-        let files = child.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = child
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
        assert_eq!(files.get("pg_logical/mappings/test1"), None);
        assert_eq!(files.get("pg_logical/mappings/test2"), None);
    }
@@ -7586,6 +7627,7 @@ mod tests {
    async fn test_metadata_image_creation() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_metadata_image_creation").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_test_timeline(TIMELINE_ID, Lsn(0x10), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -7605,8 +7647,9 @@ mod tests {
            keyspace: &KeySpace,
            lsn: Lsn,
            ctx: &RequestContext,
+            io_concurrency: IoConcurrency,
        ) -> anyhow::Result<(BTreeMap<Key, Result<Bytes, PageReconstructError>>, usize)> {
-            let mut reconstruct_state = ValuesReconstructState::default();
+            let mut reconstruct_state = ValuesReconstructState::new(io_concurrency);
            let res = tline
                .get_vectored_impl(keyspace.clone(), lsn, &mut reconstruct_state, ctx)
                .await?;
@@ -7654,7 +7697,8 @@ mod tests {

            if iter % 5 == 0 {
                let (_, before_delta_file_accessed) =
-                    scan_with_statistics(&tline, &keyspace, lsn, &ctx).await?;
+                    scan_with_statistics(&tline, &keyspace, lsn, &ctx, io_concurrency.clone())
+                        .await?;
                tline
                    .compact(
                        &cancel,
@@ -7668,7 +7712,8 @@ mod tests {
                    )
                    .await?;
                let (_, after_delta_file_accessed) =
-                    scan_with_statistics(&tline, &keyspace, lsn, &ctx).await?;
+                    scan_with_statistics(&tline, &keyspace, lsn, &ctx, io_concurrency.clone())
+                        .await?;
                assert!(after_delta_file_accessed < before_delta_file_accessed, "after_delta_file_accessed={after_delta_file_accessed}, before_delta_file_accessed={before_delta_file_accessed}");
                // Given that we already produced an image layer, there should be no delta layer needed for the scan, but still setting a low threshold there for unforeseen circumstances.
                assert!(
@@ -7757,6 +7802,7 @@ mod tests {
    async fn test_vectored_missing_metadata_key_reads() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_vectored_missing_metadata_key_reads").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let base_key = Key::from_hex("620000000033333333444444445500000000").unwrap();
        let base_key_child = Key::from_hex("620000000033333333444444445500000001").unwrap();
@@ -7895,7 +7941,7 @@ mod tests {
        );

        // test vectored scan on parent timeline
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
        let res = tline
            .get_vectored_impl(
                KeySpace::single(Key::metadata_key_range()),
@@ -7921,7 +7967,7 @@ mod tests {
        );

        // test vectored scan on child timeline
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
        let res = child
            .get_vectored_impl(
                KeySpace::single(Key::metadata_key_range()),
@@ -7959,7 +8005,9 @@ mod tests {
        lsn: Lsn,
        ctx: &RequestContext,
    ) -> Result<Option<Bytes>, GetVectoredError> {
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let io_concurrency =
+            IoConcurrency::spawn_from_conf(tline.conf, tline.gate.enter().unwrap());
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency);
        let mut res = tline
            .get_vectored_impl(
                KeySpace::single(key..key.next()),
@@ -8060,6 +8108,7 @@ mod tests {
            .await
            .unwrap();
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let key0 = Key::from_hex("620000000033333333444444445500000000").unwrap();
        let key1 = Key::from_hex("620000000033333333444444445500000001").unwrap();
@@ -8119,7 +8168,7 @@ mod tests {

        // Image layers are created at last_record_lsn
        let images = tline
-            .inspect_image_layers(Lsn(0x40), &ctx)
+            .inspect_image_layers(Lsn(0x40), &ctx, io_concurrency.clone())
            .await
            .unwrap()
            .into_iter()
@@ -8134,6 +8183,7 @@ mod tests {
            .await
            .unwrap();
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        let key1 = Key::from_hex("620000000033333333444444445500000001").unwrap();
        let key2 = Key::from_hex("620000000033333333444444445500000002").unwrap();
@@ -8184,7 +8234,7 @@ mod tests {

        // Image layers are created at last_record_lsn
        let images = tline
-            .inspect_image_layers(Lsn(0x30), &ctx)
+            .inspect_image_layers(Lsn(0x30), &ctx, io_concurrency.clone())
            .await
            .unwrap()
            .into_iter()
@@ -8197,6 +8247,7 @@ mod tests {
    async fn test_simple_bottom_most_compaction_images() -> anyhow::Result<()> {
        let harness = TenantHarness::create("test_simple_bottom_most_compaction_images").await?;
        let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();

        fn get_key(id: u32) -> Key {
            // using aux key here b/c they are guaranteed to be inside `collect_keyspace`.
@@ -8338,7 +8389,7 @@ mod tests {

        // Check if the image layer at the GC horizon contains exactly what we want
        let image_at_gc_horizon = tline
-            .inspect_image_layers(Lsn(0x30), &ctx)
+            .inspect_image_layers(Lsn(0x30), &ctx, io_concurrency.clone())
            .await
            .unwrap()
            .into_iter()
@@ -10051,7 +10102,12 @@ mod tests {

        let keyspace = KeySpace::single(get_key(0)..get_key(10));
        let results = tline
-            .get_vectored(keyspace, delta_layer_end_lsn, &ctx)
+            .get_vectored(
+                keyspace,
+                delta_layer_end_lsn,
+                IoConcurrency::sequential(),
+                &ctx,
+            )
            .await
            .expect("No vectored errors");
        for (key, res) in results {
--- a/pageserver/src/tenant/config.rs
+++ b/pageserver/src/tenant/config.rs
@@ -281,6 +281,18 @@ pub struct TenantConfOpt {
    #[serde(default)]
    pub compaction_algorithm: Option<CompactionAlgorithmSettings>,

+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_delay_threshold: Option<usize>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_stall_threshold: Option<usize>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_wait_upload: Option<bool>,
+
    #[serde(skip_serializing_if = "Option::is_none")]
    #[serde(default)]
    pub gc_horizon: Option<u64>,
@@ -357,6 +369,18 @@ pub struct TenantConfOpt {

    #[serde(skip_serializing_if = "Option::is_none")]
    pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub rel_size_v2_enabled: Option<bool>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_enabled: Option<bool>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_initial_threshold_kb: Option<u64>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_ratio_percent: Option<u64>,
 }

 impl TenantConfOpt {
@@ -382,6 +406,15 @@ impl TenantConfOpt {
                .as_ref()
                .unwrap_or(&global_conf.compaction_algorithm)
                .clone(),
+            l0_flush_delay_threshold: self
+                .l0_flush_delay_threshold
+                .or(global_conf.l0_flush_delay_threshold),
+            l0_flush_stall_threshold: self
+                .l0_flush_stall_threshold
+                .or(global_conf.l0_flush_stall_threshold),
+            l0_flush_wait_upload: self
+                .l0_flush_wait_upload
+                .unwrap_or(global_conf.l0_flush_wait_upload),
            gc_horizon: self.gc_horizon.unwrap_or(global_conf.gc_horizon),
            gc_period: self.gc_period.unwrap_or(global_conf.gc_period),
            image_creation_threshold: self
@@ -425,6 +458,16 @@ impl TenantConfOpt {
            wal_receiver_protocol_override: self
                .wal_receiver_protocol_override
                .or(global_conf.wal_receiver_protocol_override),
+            rel_size_v2_enabled: self.rel_size_v2_enabled.or(global_conf.rel_size_v2_enabled),
+            gc_compaction_enabled: self
+                .gc_compaction_enabled
+                .unwrap_or(global_conf.gc_compaction_enabled),
+            gc_compaction_initial_threshold_kb: self
+                .gc_compaction_initial_threshold_kb
+                .unwrap_or(global_conf.gc_compaction_initial_threshold_kb),
+            gc_compaction_ratio_percent: self
+                .gc_compaction_ratio_percent
+                .unwrap_or(global_conf.gc_compaction_ratio_percent),
        }
    }

@@ -436,6 +479,9 @@ impl TenantConfOpt {
            mut compaction_period,
            mut compaction_threshold,
            mut compaction_algorithm,
+            mut l0_flush_delay_threshold,
+            mut l0_flush_stall_threshold,
+            mut l0_flush_wait_upload,
            mut gc_horizon,
            mut gc_period,
            mut image_creation_threshold,
@@ -454,6 +500,10 @@ impl TenantConfOpt {
            mut lsn_lease_length_for_ts,
            mut timeline_offloading,
            mut wal_receiver_protocol_override,
+            mut rel_size_v2_enabled,
+            mut gc_compaction_enabled,
+            mut gc_compaction_initial_threshold_kb,
+            mut gc_compaction_ratio_percent,
        } = self;

        patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -470,6 +520,13 @@ impl TenantConfOpt {
            .apply(&mut compaction_period);
        patch.compaction_threshold.apply(&mut compaction_threshold);
        patch.compaction_algorithm.apply(&mut compaction_algorithm);
+        patch
+            .l0_flush_delay_threshold
+            .apply(&mut l0_flush_delay_threshold);
+        patch
+            .l0_flush_stall_threshold
+            .apply(&mut l0_flush_stall_threshold);
+        patch.l0_flush_wait_upload.apply(&mut l0_flush_wait_upload);
        patch.gc_horizon.apply(&mut gc_horizon);
        patch
            .gc_period
@@ -522,6 +579,16 @@ impl TenantConfOpt {
        patch
            .wal_receiver_protocol_override
            .apply(&mut wal_receiver_protocol_override);
+        patch.rel_size_v2_enabled.apply(&mut rel_size_v2_enabled);
+        patch
+            .gc_compaction_enabled
+            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_initial_threshold_kb
+            .apply(&mut gc_compaction_initial_threshold_kb);
+        patch
+            .gc_compaction_ratio_percent
+            .apply(&mut gc_compaction_ratio_percent);

        Ok(Self {
            checkpoint_distance,
@@ -530,6 +597,9 @@ impl TenantConfOpt {
            compaction_period,
            compaction_threshold,
            compaction_algorithm,
+            l0_flush_delay_threshold,
+            l0_flush_stall_threshold,
+            l0_flush_wait_upload,
            gc_horizon,
            gc_period,
            image_creation_threshold,
@@ -548,6 +618,10 @@ impl TenantConfOpt {
            lsn_lease_length_for_ts,
            timeline_offloading,
            wal_receiver_protocol_override,
+            rel_size_v2_enabled,
+            gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent,
        })
    }
 }
@@ -583,6 +657,9 @@ impl From<TenantConfOpt> for models::TenantConfig {
            compaction_target_size: value.compaction_target_size,
            compaction_period: value.compaction_period.map(humantime),
            compaction_threshold: value.compaction_threshold,
+            l0_flush_delay_threshold: value.l0_flush_delay_threshold,
+            l0_flush_stall_threshold: value.l0_flush_stall_threshold,
+            l0_flush_wait_upload: value.l0_flush_wait_upload,
            gc_horizon: value.gc_horizon,
            gc_period: value.gc_period.map(humantime),
            image_creation_threshold: value.image_creation_threshold,
@@ -603,6 +680,10 @@ impl From<TenantConfOpt> for models::TenantConfig {
            lsn_lease_length_for_ts: value.lsn_lease_length_for_ts.map(humantime),
            timeline_offloading: value.timeline_offloading,
            wal_receiver_protocol_override: value.wal_receiver_protocol_override,
+            rel_size_v2_enabled: value.rel_size_v2_enabled,
+            gc_compaction_enabled: value.gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb: value.gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent: value.gc_compaction_ratio_percent,
        }
    }
 }
--- a/pageserver/src/tenant/layer_map.rs
+++ b/pageserver/src/tenant/layer_map.rs
@@ -57,6 +57,7 @@ use std::collections::{HashMap, VecDeque};
 use std::iter::Peekable;
 use std::ops::Range;
 use std::sync::Arc;
+use tokio::sync::watch;
 use utils::lsn::Lsn;

 use historic_layer_coverage::BufferedHistoricLayerCoverage;
@@ -67,7 +68,6 @@ use super::storage_layer::{LayerVisibilityHint, PersistentLayerDesc};
 ///
 /// LayerMap tracks what layers exist on a timeline.
 ///
-#[derive(Default)]
 pub struct LayerMap {
    //
    // 'open_layer' holds the current InMemoryLayer that is accepting new
@@ -93,7 +93,25 @@ pub struct LayerMap {

    /// L0 layers have key range Key::MIN..Key::MAX, and locating them using R-Tree search is very inefficient.
    /// So L0 layers are held in l0_delta_layers vector, in addition to the R-tree.
+    ///
+    /// NB: make sure to notify `watch_l0_deltas` on changes.
    l0_delta_layers: Vec<Arc<PersistentLayerDesc>>,
+
+    /// Notifies about L0 delta layer changes, sending the current number of L0 layers.
+    watch_l0_deltas: watch::Sender<usize>,
+}
+
+impl Default for LayerMap {
+    fn default() -> Self {
+        Self {
+            open_layer: Default::default(),
+            next_open_layer_at: Default::default(),
+            frozen_layers: Default::default(),
+            historic: Default::default(),
+            l0_delta_layers: Default::default(),
+            watch_l0_deltas: watch::channel(0).0,
+        }
+    }
 }

 /// The primary update API for the layer map.
@@ -466,6 +484,8 @@ impl LayerMap {

        if Self::is_l0(&layer_desc.key_range, layer_desc.is_delta) {
            self.l0_delta_layers.push(layer_desc.clone().into());
+            self.watch_l0_deltas
+                .send_replace(self.l0_delta_layers.len());
        }

        self.historic.insert(
@@ -488,6 +508,8 @@ impl LayerMap {
            let mut l0_delta_layers = std::mem::take(&mut self.l0_delta_layers);
            l0_delta_layers.retain(|other| other.key() != layer_key);
            self.l0_delta_layers = l0_delta_layers;
+            self.watch_l0_deltas
+                .send_replace(self.l0_delta_layers.len());
            // this assertion is related to use of Arc::ptr_eq in Self::compare_arced_layers,
            // there's a chance that the comparison fails at runtime due to it comparing (pointer,
            // vtable) pairs.
@@ -850,6 +872,11 @@ impl LayerMap {
        &self.l0_delta_layers
    }

+    /// Subscribes to L0 delta layer changes, sending the current number of L0 delta layers.
+    pub fn watch_level0_deltas(&self) -> watch::Receiver<usize> {
+        self.watch_l0_deltas.subscribe()
+    }
+
    /// debugging function to print out the contents of the layer map
    #[allow(unused)]
    pub async fn dump(&self, verbose: bool, ctx: &RequestContext) -> Result<()> {
--- a/pageserver/src/tenant/mgr.rs
+++ b/pageserver/src/tenant/mgr.rs
@@ -1643,6 +1643,7 @@ impl TenantManager {
                        .wait_lsn(
                            *target_lsn,
                            crate::tenant::timeline::WaitLsnWaiter::Tenant,
+                            crate::tenant::timeline::WaitLsnTimeout::Default,
                            ctx,
                        )
                        .await
--- a/pageserver/src/tenant/remote_timeline_client/index.rs
+++ b/pageserver/src/tenant/remote_timeline_client/index.rs
@@ -77,8 +77,32 @@ pub struct IndexPart {
    ///
    /// None means no aux files have been written to the storage before the point
    /// when this flag is introduced.
+    ///
+    /// This flag is not used any more as all tenants have been transitioned to the new aux file policy.
    #[serde(skip_serializing_if = "Option::is_none", default)]
    pub(crate) last_aux_file_policy: Option<AuxFilePolicy>,
+
+    #[serde(skip_serializing_if = "Option::is_none", default)]
+    pub(crate) rel_size_migration: Option<RelSizeMigration>,
+
+    /// The LSN of gc-compaction horizon. Once gc-compaction is finished for all layer files below an LSN, this LSN will be updated.
+    #[serde(skip_serializing_if = "Option::is_none", default)]
+    pub(crate) l2_lsn: Option<Lsn>,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub enum RelSizeMigration {
+    /// The tenant is using the old rel_size format.
+    /// Note that this enum is persisted as `Option<RelSizeMigration>` in the index part, so
+    /// `None` is the same as `Some(RelSizeMigration::Legacy)`.
+    Legacy,
+    /// The tenant is migrating to the new rel_size format. Both old and new rel_size format are
+    /// persisted in the index part. The read path will read both formats and merge them.
+    Migrating,
+    /// The tenant has migrated to the new rel_size format. Only the new rel_size format is persisted
+    /// in the index part, and the read path will not read the old format.
+    Migrated,
 }

 impl IndexPart {
@@ -97,10 +121,12 @@ impl IndexPart {
    /// - 8: added `archived_at`
    /// - 9: +gc_blocking
    /// - 10: +import_pgdata
-    const LATEST_VERSION: usize = 10;
+    /// - 11: +rel_size_migration
+    /// - 12: +l2_lsn
+    const LATEST_VERSION: usize = 12;

    // Versions we may see when reading from a bucket.
-    pub const KNOWN_VERSIONS: &'static [usize] = &[1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
+    pub const KNOWN_VERSIONS: &'static [usize] = &[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12];

    pub const FILE_NAME: &'static str = "index_part.json";

@@ -116,6 +142,8 @@ impl IndexPart {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        }
    }

@@ -194,6 +222,10 @@ impl LayerFileMetadata {
            shard,
        }
    }
+    /// Helper to get both generation and file size in a tuple
+    pub fn generation_file_size(&self) -> (Generation, u64) {
+        (self.generation, self.file_size)
+    }
 }

 /// Limited history of earlier ancestors.
@@ -416,6 +448,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -461,6 +495,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -507,6 +543,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -556,6 +594,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let empty_layers_parsed = IndexPart::from_json_bytes(empty_layers_json.as_bytes()).unwrap();
@@ -600,6 +640,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -647,6 +689,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -699,6 +743,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: Some(AuxFilePolicy::V2),
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -756,6 +802,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: Default::default(),
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -814,6 +862,8 @@ mod tests {
            gc_blocking: None,
            last_aux_file_policy: Default::default(),
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -877,6 +927,8 @@ mod tests {
            last_aux_file_policy: Default::default(),
            archived_at: None,
            import_pgdata: None,
+            rel_size_migration: None,
+            l2_lsn: None,
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -952,7 +1004,168 @@ mod tests {
                started_at: parse_naive_datetime("2024-11-13T09:23:42.123000000"),
                finished_at: parse_naive_datetime("2024-11-13T09:42:23.123000000"),
                idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
-            })))
+            }))),
+            rel_size_migration: None,
+            l2_lsn: None,
+        };
+
+        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
+        assert_eq!(part, expected);
+    }
+
+    #[test]
+    fn v11_rel_size_migration_is_parsed() {
+        let example = r#"{
+            "version": 11,
+            "layer_metadata":{
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9": { "file_size": 25600000 },
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51": { "file_size": 9007199254741001 }
+            },
+            "disk_consistent_lsn":"0/16960E8",
+            "metadata": {
+                "disk_consistent_lsn": "0/16960E8",
+                "prev_record_lsn": "0/1696070",
+                "ancestor_timeline": "e45a7f37d3ee2ff17dc14bf4f4e3f52e",
+                "ancestor_lsn": "0/0",
+                "latest_gc_cutoff_lsn": "0/1696070",
+                "initdb_lsn": "0/1696070",
+                "pg_version": 14
+            },
+            "gc_blocking": {
+                "started_at": "2024-07-19T09:00:00.123",
+                "reasons": ["DetachAncestor"]
+            },
+            "import_pgdata": {
+                "V1": {
+                    "Done": {
+                        "idempotency_key": "specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5",
+                        "started_at": "2024-11-13T09:23:42.123",
+                        "finished_at": "2024-11-13T09:42:23.123"
+                    }
+                }
+            },
+            "rel_size_migration": "legacy"
+        }"#;
+
+        let expected = IndexPart {
+            version: 11,
+            layer_metadata: HashMap::from([
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9".parse().unwrap(), LayerFileMetadata {
+                    file_size: 25600000,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                }),
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51".parse().unwrap(), LayerFileMetadata {
+                    file_size: 9007199254741001,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                })
+            ]),
+            disk_consistent_lsn: "0/16960E8".parse::<Lsn>().unwrap(),
+            metadata: TimelineMetadata::new(
+                Lsn::from_str("0/16960E8").unwrap(),
+                Some(Lsn::from_str("0/1696070").unwrap()),
+                Some(TimelineId::from_str("e45a7f37d3ee2ff17dc14bf4f4e3f52e").unwrap()),
+                Lsn::INVALID,
+                Lsn::from_str("0/1696070").unwrap(),
+                Lsn::from_str("0/1696070").unwrap(),
+                14,
+            ).with_recalculated_checksum().unwrap(),
+            deleted_at: None,
+            lineage: Default::default(),
+            gc_blocking: Some(GcBlocking {
+                started_at: parse_naive_datetime("2024-07-19T09:00:00.123000000"),
+                reasons: enumset::EnumSet::from_iter([GcBlockingReason::DetachAncestor]),
+            }),
+            last_aux_file_policy: Default::default(),
+            archived_at: None,
+            import_pgdata: Some(import_pgdata::index_part_format::Root::V1(import_pgdata::index_part_format::V1::Done(import_pgdata::index_part_format::Done{
+                started_at: parse_naive_datetime("2024-11-13T09:23:42.123000000"),
+                finished_at: parse_naive_datetime("2024-11-13T09:42:23.123000000"),
+                idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
+            }))),
+            rel_size_migration: Some(RelSizeMigration::Legacy),
+            l2_lsn: None,
+        };
+
+        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
+        assert_eq!(part, expected);
+    }
+
+    #[test]
+    fn v12_l2_lsn_is_parsed() {
+        let example = r#"{
+            "version": 12,
+            "layer_metadata":{
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9": { "file_size": 25600000 },
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51": { "file_size": 9007199254741001 }
+            },
+            "disk_consistent_lsn":"0/16960E8",
+            "metadata": {
+                "disk_consistent_lsn": "0/16960E8",
+                "prev_record_lsn": "0/1696070",
+                "ancestor_timeline": "e45a7f37d3ee2ff17dc14bf4f4e3f52e",
+                "ancestor_lsn": "0/0",
+                "latest_gc_cutoff_lsn": "0/1696070",
+                "initdb_lsn": "0/1696070",
+                "pg_version": 14
+            },
+            "gc_blocking": {
+                "started_at": "2024-07-19T09:00:00.123",
+                "reasons": ["DetachAncestor"]
+            },
+            "import_pgdata": {
+                "V1": {
+                    "Done": {
+                        "idempotency_key": "specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5",
+                        "started_at": "2024-11-13T09:23:42.123",
+                        "finished_at": "2024-11-13T09:42:23.123"
+                    }
+                }
+            },
+            "rel_size_migration": "legacy",
+            "l2_lsn": "0/16960E8"
+        }"#;
+
+        let expected = IndexPart {
+            version: 12,
+            layer_metadata: HashMap::from([
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9".parse().unwrap(), LayerFileMetadata {
+                    file_size: 25600000,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                }),
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51".parse().unwrap(), LayerFileMetadata {
+                    file_size: 9007199254741001,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                })
+            ]),
+            disk_consistent_lsn: "0/16960E8".parse::<Lsn>().unwrap(),
+            metadata: TimelineMetadata::new(
+                Lsn::from_str("0/16960E8").unwrap(),
+                Some(Lsn::from_str("0/1696070").unwrap()),
+                Some(TimelineId::from_str("e45a7f37d3ee2ff17dc14bf4f4e3f52e").unwrap()),
+                Lsn::INVALID,
+                Lsn::from_str("0/1696070").unwrap(),
+                Lsn::from_str("0/1696070").unwrap(),
+                14,
+            ).with_recalculated_checksum().unwrap(),
+            deleted_at: None,
+            lineage: Default::default(),
+            gc_blocking: Some(GcBlocking {
+                started_at: parse_naive_datetime("2024-07-19T09:00:00.123000000"),
+                reasons: enumset::EnumSet::from_iter([GcBlockingReason::DetachAncestor]),
+            }),
+            last_aux_file_policy: Default::default(),
+            archived_at: None,
+            import_pgdata: Some(import_pgdata::index_part_format::Root::V1(import_pgdata::index_part_format::V1::Done(import_pgdata::index_part_format::Done{
+                started_at: parse_naive_datetime("2024-11-13T09:23:42.123000000"),
+                finished_at: parse_naive_datetime("2024-11-13T09:42:23.123000000"),
+                idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
+            }))),
+            rel_size_migration: Some(RelSizeMigration::Legacy),
+            l2_lsn: Some("0/16960E8".parse::<Lsn>().unwrap()),
        };

        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
--- a/pageserver/src/tenant/secondary/downloader.rs
+++ b/pageserver/src/tenant/secondary/downloader.rs
@@ -559,6 +559,13 @@ impl JobGenerator<PendingDownload, RunningDownload, CompleteDownload, DownloadCo
    }
 }

+enum LayerAction {
+    Download,
+    NoAction,
+    Skip,
+    Touch,
+}
+
 /// This type is a convenience to group together the various functions involved in
 /// freshening a secondary tenant.
 struct TenantDownloader<'a> {
@@ -1008,69 +1015,17 @@ impl<'a> TenantDownloader<'a> {
                return (Err(UpdateError::Restart), touched);
            }

-            // Existing on-disk layers: just update their access time.
-            if let Some(on_disk) = timeline_state.on_disk_layers.get(&layer.name) {
-                tracing::debug!("Layer {} is already on disk", layer.name);
-
-                if cfg!(debug_assertions) {
-                    // Debug for https://github.com/neondatabase/neon/issues/6966: check that the files we think
-                    // are already present on disk are really there.
-                    match tokio::fs::metadata(&on_disk.local_path).await {
-                        Ok(meta) => {
-                            tracing::debug!(
-                                "Layer {} present at {}, size {}",
-                                layer.name,
-                                on_disk.local_path,
-                                meta.len(),
-                            );
-                        }
-                        Err(e) => {
-                            tracing::warn!(
-                                "Layer {} not found at {} ({})",
-                                layer.name,
-                                on_disk.local_path,
-                                e
-                            );
-                            debug_assert!(false);
-                        }
-                    }
-                }
-
-                if on_disk.metadata != layer.metadata || on_disk.access_time != layer.access_time {
-                    // We already have this layer on disk.  Update its access time.
-                    tracing::debug!(
-                        "Access time updated for layer {}: {} -> {}",
-                        layer.name,
-                        strftime(&on_disk.access_time),
-                        strftime(&layer.access_time)
-                    );
-                    touched.push(layer);
-                }
-                continue;
-            } else {
-                tracing::debug!("Layer {} not present on disk yet", layer.name);
-            }
-
-            // Eviction: if we evicted a layer, then do not re-download it unless it was accessed more
-            // recently than it was evicted.
-            if let Some(evicted_at) = timeline_state.evicted_at.get(&layer.name) {
-                if &layer.access_time > evicted_at {
-                    tracing::info!(
-                        "Re-downloading evicted layer {}, accessed at {}, evicted at {}",
-                        layer.name,
-                        strftime(&layer.access_time),
-                        strftime(evicted_at)
-                    );
-                } else {
-                    tracing::trace!(
-                        "Not re-downloading evicted layer {}, accessed at {}, evicted at {}",
-                        layer.name,
-                        strftime(&layer.access_time),
-                        strftime(evicted_at)
-                    );
+            match self.layer_action(&timeline_state, &layer).await {
+                LayerAction::Download => (),
+                LayerAction::NoAction => continue,
+                LayerAction::Skip => {
                    self.skip_layer(layer);
                    continue;
                }
+                LayerAction::Touch => {
+                    touched.push(layer);
+                    continue;
+                }
            }

            match self
@@ -1091,6 +1046,86 @@ impl<'a> TenantDownloader<'a> {
        (Ok(()), touched)
    }

+    async fn layer_action(
+        &self,
+        timeline_state: &SecondaryDetailTimeline,
+        layer: &HeatMapLayer,
+    ) -> LayerAction {
+        // Existing on-disk layers: just update their access time.
+        if let Some(on_disk) = timeline_state.on_disk_layers.get(&layer.name) {
+            tracing::debug!("Layer {} is already on disk", layer.name);
+
+            if cfg!(debug_assertions) {
+                // Debug for https://github.com/neondatabase/neon/issues/6966: check that the files we think
+                // are already present on disk are really there.
+                match tokio::fs::metadata(&on_disk.local_path).await {
+                    Ok(meta) => {
+                        tracing::debug!(
+                            "Layer {} present at {}, size {}",
+                            layer.name,
+                            on_disk.local_path,
+                            meta.len(),
+                        );
+                    }
+                    Err(e) => {
+                        tracing::warn!(
+                            "Layer {} not found at {} ({})",
+                            layer.name,
+                            on_disk.local_path,
+                            e
+                        );
+                        debug_assert!(false);
+                    }
+                }
+            }
+
+            if on_disk.metadata.generation_file_size() != on_disk.metadata.generation_file_size() {
+                tracing::info!(
+                    "Re-downloading layer {} with changed size or generation: {:?}->{:?}",
+                    layer.name,
+                    on_disk.metadata.generation_file_size(),
+                    on_disk.metadata.generation_file_size()
+                );
+                return LayerAction::Download;
+            }
+            if on_disk.metadata != layer.metadata || on_disk.access_time != layer.access_time {
+                // We already have this layer on disk.  Update its access time.
+                tracing::debug!(
+                    "Access time updated for layer {}: {} -> {}",
+                    layer.name,
+                    strftime(&on_disk.access_time),
+                    strftime(&layer.access_time)
+                );
+                return LayerAction::Touch;
+            }
+            return LayerAction::NoAction;
+        } else {
+            tracing::debug!("Layer {} not present on disk yet", layer.name);
+        }
+
+        // Eviction: if we evicted a layer, then do not re-download it unless it was accessed more
+        // recently than it was evicted.
+        if let Some(evicted_at) = timeline_state.evicted_at.get(&layer.name) {
+            if &layer.access_time > evicted_at {
+                tracing::info!(
+                    "Re-downloading evicted layer {}, accessed at {}, evicted at {}",
+                    layer.name,
+                    strftime(&layer.access_time),
+                    strftime(evicted_at)
+                );
+            } else {
+                tracing::trace!(
+                    "Not re-downloading evicted layer {}, accessed at {}, evicted at {}",
+                    layer.name,
+                    strftime(&layer.access_time),
+                    strftime(evicted_at)
+                );
+                return LayerAction::Skip;
+            }
+        }
+        LayerAction::Download
+    }
+
    async fn download_timeline(
        &self,
        timeline: HeatMapTimeline,
--- a/pageserver/src/tenant/storage_layer.rs
+++ b/pageserver/src/tenant/storage_layer.rs
@@ -10,18 +10,26 @@ mod layer_desc;
 mod layer_name;
 pub mod merge_iterator;

+use crate::config::PageServerConf;
 use crate::context::{AccessStatsBehavior, RequestContext};
 use bytes::Bytes;
+use futures::stream::FuturesUnordered;
+use futures::StreamExt;
 use pageserver_api::key::Key;
 use pageserver_api::keyspace::{KeySpace, KeySpaceRandomAccum};
 use pageserver_api::record::NeonWalRecord;
 use pageserver_api::value::Value;
-use std::cmp::{Ordering, Reverse};
+use std::cmp::Ordering;
 use std::collections::hash_map::Entry;
 use std::collections::{BinaryHeap, HashMap};
+use std::future::Future;
 use std::ops::Range;
+use std::pin::Pin;
+use std::sync::atomic::AtomicUsize;
 use std::sync::Arc;
 use std::time::{Duration, SystemTime, UNIX_EPOCH};
+use tracing::{trace, Instrument};
+use utils::sync::gate::GateGuard;

 use utils::lsn::Lsn;

@@ -78,30 +86,151 @@ pub(crate) enum ValueReconstructSituation {
    Continue,
 }

-/// Reconstruct data accumulated for a single key during a vectored get
-#[derive(Debug, Default, Clone)]
-pub(crate) struct VectoredValueReconstructState {
-    pub(crate) records: Vec<(Lsn, NeonWalRecord)>,
-    pub(crate) img: Option<(Lsn, Bytes)>,
-
-    situation: ValueReconstructSituation,
+/// On disk representation of a value loaded in a buffer
+#[derive(Debug)]
+pub(crate) enum OnDiskValue {
+    /// Unencoded [`Value::Image`]
+    RawImage(Bytes),
+    /// Encoded [`Value`]. Can deserialize into an image or a WAL record
+    WalRecordOrImage(Bytes),
 }

-impl VectoredValueReconstructState {
-    fn get_cached_lsn(&self) -> Option<Lsn> {
-        self.img.as_ref().map(|img| img.0)
+/// Reconstruct data accumulated for a single key during a vectored get
+#[derive(Debug, Default)]
+pub(crate) struct VectoredValueReconstructState {
+    pub(crate) on_disk_values: Vec<(Lsn, OnDiskValueIoWaiter)>,
+
+    pub(crate) situation: ValueReconstructSituation,
+}
+
+#[derive(Debug)]
+pub(crate) struct OnDiskValueIoWaiter {
+    rx: tokio::sync::oneshot::Receiver<OnDiskValueIoResult>,
+}
+
+#[derive(Debug)]
+#[must_use]
+pub(crate) enum OnDiskValueIo {
+    /// Traversal identified this IO as required to complete the vectored get.
+    Required {
+        num_active_ios: Arc<AtomicUsize>,
+        tx: tokio::sync::oneshot::Sender<OnDiskValueIoResult>,
+    },
+    /// Sparse keyspace reads always read all the values for a given key,
+    /// even though only the first value is needed.
+    ///
+    /// This variant represents the unnecessary IOs for those values at lower LSNs
+    /// that aren't needed, but are currently still being done.
+    ///
+    /// The execution of unnecessary IOs was a pre-existing behavior before concurrent IO.
+    /// We added this explicit representation here so that we can drop
+    /// unnecessary IO results immediately, instead of buffering them in
+    /// `oneshot` channels inside [`VectoredValueReconstructState`] until
+    /// [`VectoredValueReconstructState::collect_pending_ios`] gets called.
+    Unnecessary,
+}
+
+type OnDiskValueIoResult = Result<OnDiskValue, std::io::Error>;
+
+impl OnDiskValueIo {
+    pub(crate) fn complete(self, res: OnDiskValueIoResult) {
+        match self {
+            OnDiskValueIo::Required { num_active_ios, tx } => {
+                num_active_ios.fetch_sub(1, std::sync::atomic::Ordering::Release);
+                let _ = tx.send(res);
+            }
+            OnDiskValueIo::Unnecessary => {
+                // Nobody cared, see variant doc comment.
+            }
+        }
    }
 }

-impl From<VectoredValueReconstructState> for ValueReconstructState {
-    fn from(mut state: VectoredValueReconstructState) -> Self {
-        // walredo expects the records to be descending in terms of Lsn
-        state.records.sort_by_key(|(lsn, _)| Reverse(*lsn));
+#[derive(Debug, thiserror::Error)]
+pub(crate) enum WaitCompletionError {
+    #[error("OnDiskValueIo was dropped without completing, likely the sidecar task panicked")]
+    IoDropped,
+}

-        ValueReconstructState {
-            records: state.records,
-            img: state.img,
+impl OnDiskValueIoWaiter {
+    pub(crate) async fn wait_completion(self) -> Result<OnDiskValueIoResult, WaitCompletionError> {
+        // NB: for Unnecessary IOs, this method never gets called because we don't add them to `on_disk_values`.
+        self.rx.await.map_err(|_| WaitCompletionError::IoDropped)
+    }
+}
+
+impl VectoredValueReconstructState {
+    /// # Cancel-Safety
+    ///
+    /// Technically fine to stop polling this future, but, the IOs will still
+    /// be executed to completion by the sidecar task and hold on to / consume resources.
+    /// Better not do it to make reasonsing about the system easier.
+    pub(crate) async fn collect_pending_ios(
+        self,
+    ) -> Result<ValueReconstructState, PageReconstructError> {
+        use utils::bin_ser::BeSer;
+
+        let mut res = Ok(ValueReconstructState::default());
+
+        // We should try hard not to bail early, so that by the time we return from this
+        // function, all IO for this value is done. It's not required -- we could totally
+        // stop polling the IO futures in the sidecar task, they need to support that,
+        // but just stopping to poll doesn't reduce the IO load on the disk. It's easier
+        // to reason about the system if we just wait for all IO to complete, even if
+        // we're no longer interested in the result.
+        //
+        // Revisit this when IO futures are replaced with a more sophisticated IO system
+        // and an IO scheduler, where we know which IOs were submitted and which ones
+        // just queued. Cf the comment on IoConcurrency::spawn_io.
+        for (lsn, waiter) in self.on_disk_values {
+            let value_recv_res = waiter
+                .wait_completion()
+                // we rely on the caller to poll us to completion, so this is not a bail point
+                .await;
+            // Force not bailing early by wrapping the code into a closure.
+            #[allow(clippy::redundant_closure_call)]
+            let _: () = (|| {
+                match (&mut res, value_recv_res) {
+                    (Err(_), _) => {
+                        // We've already failed, no need to process more.
+                    }
+                    (Ok(_), Err(wait_err)) => {
+                        // This shouldn't happen - likely the sidecar task panicked.
+                        res = Err(PageReconstructError::Other(wait_err.into()));
+                    }
+                    (Ok(_), Ok(Err(err))) => {
+                        let err: std::io::Error = err;
+                        // TODO: returning IO error here will fail a compute query.
+                        // Probably not what we want, we're not doing `maybe_fatal_err`
+                        // in the IO futures.
+                        // But it's been like that for a long time, not changing it
+                        // as part of concurrent IO.
+                        // => https://github.com/neondatabase/neon/issues/10454
+                        res = Err(PageReconstructError::Other(err.into()));
+                    }
+                    (Ok(ok), Ok(Ok(OnDiskValue::RawImage(img)))) => {
+                        assert!(ok.img.is_none());
+                        ok.img = Some((lsn, img));
+                    }
+                    (Ok(ok), Ok(Ok(OnDiskValue::WalRecordOrImage(buf)))) => {
+                        match Value::des(&buf) {
+                            Ok(Value::WalRecord(rec)) => {
+                                ok.records.push((lsn, rec));
+                            }
+                            Ok(Value::Image(img)) => {
+                                assert!(ok.img.is_none());
+                                ok.img = Some((lsn, img));
+                            }
+                            Err(err) => {
+                                res = Err(PageReconstructError::Other(err.into()));
+                            }
+                        }
+                    }
+                }
+            })();
        }
+
+        res
    }
 }

@@ -109,7 +238,7 @@ impl From<VectoredValueReconstructState> for ValueReconstructState {
 pub(crate) struct ValuesReconstructState {
    /// The keys will be removed after `get_vectored` completes. The caller outside `Timeline`
    /// should not expect to get anything from this hashmap.
-    pub(crate) keys: HashMap<Key, Result<VectoredValueReconstructState, PageReconstructError>>,
+    pub(crate) keys: HashMap<Key, VectoredValueReconstructState>,
    /// The keys which are already retrieved
    keys_done: KeySpaceRandomAccum,

@@ -119,27 +248,365 @@ pub(crate) struct ValuesReconstructState {
    // Statistics that are still accessible as a caller of `get_vectored_impl`.
    layers_visited: u32,
    delta_layers_visited: u32,
+
+    pub(crate) io_concurrency: IoConcurrency,
+    num_active_ios: Arc<AtomicUsize>,
+}
+
+/// The level of IO concurrency to be used on the read path
+///
+/// The desired end state is that we always do parallel IO.
+/// This struct and the dispatching in the impl will be removed once
+/// we've built enough confidence.
+pub(crate) enum IoConcurrency {
+    Sequential,
+    SidecarTask {
+        task_id: usize,
+        ios_tx: tokio::sync::mpsc::UnboundedSender<IoFuture>,
+    },
+}
+
+type IoFuture = Pin<Box<dyn Send + Future<Output = ()>>>;
+
+pub(crate) enum SelectedIoConcurrency {
+    Sequential,
+    SidecarTask(GateGuard),
+}
+
+impl std::fmt::Debug for IoConcurrency {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            IoConcurrency::Sequential => write!(f, "Sequential"),
+            IoConcurrency::SidecarTask { .. } => write!(f, "SidecarTask"),
+        }
+    }
+}
+
+impl std::fmt::Debug for SelectedIoConcurrency {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            SelectedIoConcurrency::Sequential => write!(f, "Sequential"),
+            SelectedIoConcurrency::SidecarTask(_) => write!(f, "SidecarTask"),
+        }
+    }
+}
+
+impl IoConcurrency {
+    /// Force sequential IO. This is a temporary workaround until we have
+    /// moved plumbing-through-the-call-stack
+    /// of IoConcurrency into `RequestContextq.
+    ///
+    /// DO NOT USE for new code.
+    ///
+    /// Tracking issue: <https://github.com/neondatabase/neon/issues/10460>.
+    pub(crate) fn sequential() -> Self {
+        Self::spawn(SelectedIoConcurrency::Sequential)
+    }
+
+    pub(crate) fn spawn_from_conf(
+        conf: &'static PageServerConf,
+        gate_guard: GateGuard,
+    ) -> IoConcurrency {
+        use pageserver_api::config::GetVectoredConcurrentIo;
+        let selected = match conf.get_vectored_concurrent_io {
+            GetVectoredConcurrentIo::Sequential => SelectedIoConcurrency::Sequential,
+            GetVectoredConcurrentIo::SidecarTask => SelectedIoConcurrency::SidecarTask(gate_guard),
+        };
+        Self::spawn(selected)
+    }
+
+    pub(crate) fn spawn(io_concurrency: SelectedIoConcurrency) -> Self {
+        match io_concurrency {
+            SelectedIoConcurrency::Sequential => IoConcurrency::Sequential,
+            SelectedIoConcurrency::SidecarTask(gate_guard) => {
+                let (ios_tx, ios_rx) = tokio::sync::mpsc::unbounded_channel();
+                static TASK_ID: AtomicUsize = AtomicUsize::new(0);
+                let task_id = TASK_ID.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+                // TODO: enrich the span with more context (tenant,shard,timeline) + (basebackup|pagestream|...)
+                let span =
+                    tracing::info_span!(parent: None, "IoConcurrency_sidecar", task_id = task_id);
+                trace!(task_id, "spawning sidecar task");
+                tokio::spawn(async move {
+                    trace!("start");
+                    scopeguard::defer!{ trace!("end") };
+                    type IosRx = tokio::sync::mpsc::UnboundedReceiver<IoFuture>;
+                    enum State {
+                        Waiting {
+                            // invariant: is_empty(), but we recycle the allocation
+                            empty_futures: FuturesUnordered<IoFuture>,
+                            ios_rx: IosRx,
+                        },
+                        Executing {
+                            futures: FuturesUnordered<IoFuture>,
+                            ios_rx: IosRx,
+                        },
+                        ShuttingDown {
+                            futures: FuturesUnordered<IoFuture>,
+                        },
+                    }
+                    let mut state = State::Waiting {
+                        empty_futures: FuturesUnordered::new(),
+                        ios_rx,
+                    };
+                    loop {
+                        match state {
+                            State::Waiting {
+                                empty_futures,
+                                mut ios_rx,
+                            } => {
+                                assert!(empty_futures.is_empty());
+                                tokio::select! {
+                                    fut = ios_rx.recv() => {
+                                        if let Some(fut) = fut {
+                                            trace!("received new io future");
+                                            empty_futures.push(fut);
+                                            state = State::Executing { futures: empty_futures, ios_rx };
+                                        } else {
+                                            state = State::ShuttingDown { futures: empty_futures }
+                                        }
+                                    }
+                                }
+                            }
+                            State::Executing {
+                                mut futures,
+                                mut ios_rx,
+                            } => {
+                                tokio::select! {
+                                    res = futures.next() => {
+                                        trace!("io future completed");
+                                        assert!(res.is_some());
+                                        if futures.is_empty() {
+                                            state = State::Waiting { empty_futures: futures, ios_rx};
+                                        } else {
+                                            state = State::Executing { futures, ios_rx };
+                                        }
+                                    }
+                                    fut = ios_rx.recv() => {
+                                        if let Some(fut) = fut {
+                                            trace!("received new io future");
+                                            futures.push(fut);
+                                            state =  State::Executing { futures, ios_rx};
+                                        } else {
+                                            state = State::ShuttingDown { futures };
+                                        }
+                                    }
+                                }
+                            }
+                            State::ShuttingDown {
+                                mut futures,
+                            } => {
+                                trace!("shutting down");
+                                while let Some(()) = futures.next().await {
+                                    trace!("io future completed (shutdown)");
+                                    // drain
+                                }
+                                trace!("shutdown complete");
+                                break;
+                            }
+                        }
+                    }
+                    drop(gate_guard); // drop it right before we exit
+                }.instrument(span));
+                IoConcurrency::SidecarTask { task_id, ios_tx }
+            }
+        }
+    }
+
+    pub(crate) fn clone(&self) -> Self {
+        match self {
+            IoConcurrency::Sequential => IoConcurrency::Sequential,
+            IoConcurrency::SidecarTask { task_id, ios_tx } => IoConcurrency::SidecarTask {
+                task_id: *task_id,
+                ios_tx: ios_tx.clone(),
+            },
+        }
+    }
+
+    /// Submit an IO to be executed in the background. DEADLOCK RISK, read the full doc string.
+    ///
+    /// The IO is represented as an opaque future.
+    /// IO completion must be handled inside the future, e.g., through a oneshot channel.
+    ///
+    /// The API seems simple but there are multiple **pitfalls** involving
+    /// DEADLOCK RISK.
+    ///
+    /// First, there are no guarantees about the exexecution of the IO.
+    /// It may be `await`ed in-place before this function returns.
+    /// It may be polled partially by this task and handed off to another task to be finished.
+    /// It may be polled and then dropped before returning ready.
+    ///
+    /// This means that submitted IOs must not be interedependent.
+    /// Interdependence may be through shared limited resources, e.g.,
+    /// - VirtualFile file descriptor cache slot acquisition
+    /// - tokio-epoll-uring slot
+    ///
+    /// # Why current usage is safe from deadlocks
+    ///
+    /// Textbook condition for a deadlock is that _all_ of the following be given
+    /// - Mutual exclusion
+    /// - Hold and wait
+    /// - No preemption
+    /// - Circular wait
+    ///
+    /// The current usage is safe because:
+    /// - Mutual exclusion: IO futures definitely use mutexes, no way around that for now
+    /// - Hold and wait: IO futures currently hold two kinds of locks/resources while waiting
+    ///   for acquisition of other resources:
+    ///    - VirtualFile file descriptor cache slot tokio mutex
+    ///    - tokio-epoll-uring slot (uses tokio notify => wait queue, much like mutex)
+    /// - No preemption: there's no taking-away of acquired locks/resources => given
+    /// - Circular wait: this is the part of the condition that isn't met: all IO futures
+    ///   first acquire VirtualFile mutex, then tokio-epoll-uring slot.
+    ///   There is no IO future that acquires slot before VirtualFile.
+    ///   Hence there can be no circular waiting.
+    ///   Hence there cannot be a deadlock.
+    ///
+    /// This is a very fragile situation and must be revisited whenver any code called from
+    /// inside the IO futures is changed.
+    ///
+    /// We will move away from opaque IO futures towards well-defined IOs at some point in
+    /// the future when we have shipped this first version of concurrent IO to production
+    /// and are ready to retire the Sequential mode which runs the futures in place.
+    /// Right now, while brittle, the opaque IO approach allows us to ship the feature
+    /// with minimal changes to the code and minimal changes to existing behavior in Sequential mode.
+    ///
+    /// Also read the comment in `collect_pending_ios`.
+    pub(crate) async fn spawn_io<F>(&mut self, fut: F)
+    where
+        F: std::future::Future<Output = ()> + Send + 'static,
+    {
+        match self {
+            IoConcurrency::Sequential => fut.await,
+            IoConcurrency::SidecarTask { ios_tx, .. } => {
+                let fut = Box::pin(fut);
+                // NB: experiments showed that doing an opportunistic poll of `fut` here was bad for throughput
+                // while insignificant for latency.
+                // It would make sense to revisit the tokio-epoll-uring API in the future such that we can try
+                // a submission here, but never poll the future. That way, io_uring can make proccess while
+                // the future sits in the ios_tx queue.
+                match ios_tx.send(fut) {
+                    Ok(()) => {}
+                    Err(_) => {
+                        unreachable!("the io task must have exited, likely it panicked")
+                    }
+                }
+            }
+        }
+    }
+
+    #[cfg(test)]
+    pub(crate) fn spawn_for_test() -> impl std::ops::DerefMut<Target = Self> {
+        use std::ops::{Deref, DerefMut};
+        use tracing::info;
+        use utils::sync::gate::Gate;
+
+        // Spawn needs a Gate, give it one.
+        struct Wrapper {
+            inner: IoConcurrency,
+            #[allow(dead_code)]
+            gate: Box<Gate>,
+        }
+        impl Deref for Wrapper {
+            type Target = IoConcurrency;
+
+            fn deref(&self) -> &Self::Target {
+                &self.inner
+            }
+        }
+        impl DerefMut for Wrapper {
+            fn deref_mut(&mut self) -> &mut Self::Target {
+                &mut self.inner
+            }
+        }
+        let gate = Box::new(Gate::default());
+
+        // The default behavior when running Rust unit tests without any further
+        // flags is to use the new behavior.
+        // The CI uses the following environment variable to unit test both old
+        // and new behavior.
+        // NB: the Python regression & perf tests take the `else` branch
+        // below and have their own defaults management.
+        let selected = {
+            // The pageserver_api::config type is unsuitable because it's internally tagged.
+            #[derive(serde::Deserialize)]
+            #[serde(rename_all = "kebab-case")]
+            enum TestOverride {
+                Sequential,
+                SidecarTask,
+            }
+            use once_cell::sync::Lazy;
+            static TEST_OVERRIDE: Lazy<TestOverride> = Lazy::new(|| {
+                utils::env::var_serde_json_string(
+                    "NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO",
+                )
+                .unwrap_or(TestOverride::SidecarTask)
+            });
+
+            match *TEST_OVERRIDE {
+                TestOverride::Sequential => SelectedIoConcurrency::Sequential,
+                TestOverride::SidecarTask => {
+                    SelectedIoConcurrency::SidecarTask(gate.enter().expect("just created it"))
+                }
+            }
+        };
+
+        info!(?selected, "get_vectored_concurrent_io test");
+
+        Wrapper {
+            inner: Self::spawn(selected),
+            gate,
+        }
+    }
+}
+
+/// Make noise in case the [`ValuesReconstructState`] gets dropped while
+/// there are still IOs in flight.
+/// Refer to `collect_pending_ios` for why we prefer not to do that.
+//
+/// We log from here instead of from the sidecar task because the [`ValuesReconstructState`]
+/// gets dropped in a tracing span with more context.
+/// We repeat the sidecar tasks's `task_id` so we can correlate what we emit here with
+/// the logs / panic handler logs from the sidecar task, which also logs the `task_id`.
+impl Drop for ValuesReconstructState {
+    fn drop(&mut self) {
+        let num_active_ios = self
+            .num_active_ios
+            .load(std::sync::atomic::Ordering::Acquire);
+        if num_active_ios == 0 {
+            return;
+        }
+        let sidecar_task_id = match &self.io_concurrency {
+            IoConcurrency::Sequential => None,
+            IoConcurrency::SidecarTask { task_id, .. } => Some(*task_id),
+        };
+        tracing::warn!(
+            num_active_ios,
+            ?sidecar_task_id,
+            backtrace=%std::backtrace::Backtrace::force_capture(),
+            "dropping ValuesReconstructState while some IOs have not been completed",
+        );
+    }
 }

 impl ValuesReconstructState {
-    pub(crate) fn new() -> Self {
+    pub(crate) fn new(io_concurrency: IoConcurrency) -> Self {
        Self {
            keys: HashMap::new(),
            keys_done: KeySpaceRandomAccum::new(),
            keys_with_image_coverage: None,
            layers_visited: 0,
            delta_layers_visited: 0,
+            io_concurrency,
+            num_active_ios: Arc::new(AtomicUsize::new(0)),
        }
    }

-    /// Associate a key with the error which it encountered and mark it as done
-    pub(crate) fn on_key_error(&mut self, key: Key, err: PageReconstructError) {
-        let previous = self.keys.insert(key, Err(err));
-        if let Some(Ok(state)) = previous {
-            if state.situation == ValueReconstructSituation::Continue {
-                self.keys_done.add_key(key);
-            }
-        }
+    /// Absolutely read [`IoConcurrency::spawn_io`] to learn about assumptions & pitfalls.
+    pub(crate) async fn spawn_io<F>(&mut self, fut: F)
+    where
+        F: std::future::Future<Output = ()> + Send + 'static,
+    {
+        self.io_concurrency.spawn_io(fut).await;
    }

    pub(crate) fn on_layer_visited(&mut self, layer: &ReadableLayer) {
@@ -159,29 +626,6 @@ impl ValuesReconstructState {
        self.layers_visited
    }

-    /// This function is called after reading a keyspace from a layer.
-    /// It checks if the read path has now moved past the cached Lsn for any keys.
-    ///
-    /// Implementation note: We intentionally iterate over the keys for which we've
-    /// already collected some reconstruct data. This avoids scaling complexity with
-    /// the size of the search space.
-    pub(crate) fn on_lsn_advanced(&mut self, keyspace: &KeySpace, advanced_to: Lsn) {
-        for (key, value) in self.keys.iter_mut() {
-            if !keyspace.contains(key) {
-                continue;
-            }
-
-            if let Ok(state) = value {
-                if state.situation != ValueReconstructSituation::Complete
-                    && state.get_cached_lsn() >= Some(advanced_to)
-                {
-                    state.situation = ValueReconstructSituation::Complete;
-                    self.keys_done.add_key(*key);
-                }
-            }
-        }
-    }
-
    /// On hitting image layer, we can mark all keys in this range as done, because
    /// if the image layer does not contain a key, it is deleted/never added.
    pub(crate) fn on_image_layer_visited(&mut self, key_range: &Range<Key>) {
@@ -199,70 +643,42 @@ impl ValuesReconstructState {
    ///
    /// If the key is in the sparse keyspace (i.e., aux files), we do not track them in
    /// `key_done`.
-    pub(crate) fn update_key(
-        &mut self,
-        key: &Key,
-        lsn: Lsn,
-        value: Value,
-    ) -> ValueReconstructSituation {
-        let state = self
-            .keys
-            .entry(*key)
-            .or_insert(Ok(VectoredValueReconstructState::default()));
+    // TODO: rename this method & update description.
+    pub(crate) fn update_key(&mut self, key: &Key, lsn: Lsn, completes: bool) -> OnDiskValueIo {
+        let state = self.keys.entry(*key).or_default();
+
        let is_sparse_key = key.is_sparse();
-        if let Ok(state) = state {
-            let key_done = match state.situation {
-                ValueReconstructSituation::Complete => {
-                    if is_sparse_key {
-                        // Sparse keyspace might be visited multiple times because
-                        // we don't track unmapped keyspaces.
-                        return ValueReconstructSituation::Complete;
-                    } else {
-                        unreachable!()
-                    }
-                }
-                ValueReconstructSituation::Continue => match value {
-                    Value::Image(img) => {
-                        state.img = Some((lsn, img));
-                        true
-                    }
-                    Value::WalRecord(rec) => {
-                        debug_assert!(
-                            Some(lsn) > state.get_cached_lsn(),
-                            "Attempt to collect a record below cached LSN for walredo: {} < {}",
-                            lsn,
-                            state
-                                .get_cached_lsn()
-                                .expect("Assertion can only fire if a cached lsn is present")
-                        );

-                        let will_init = rec.will_init();
-                        state.records.push((lsn, rec));
-                        will_init
-                    }
-                },
-            };
-
-            if key_done && state.situation == ValueReconstructSituation::Continue {
-                state.situation = ValueReconstructSituation::Complete;
-                if !is_sparse_key {
-                    self.keys_done.add_key(*key);
+        let required_io = match state.situation {
+            ValueReconstructSituation::Complete => {
+                if is_sparse_key {
+                    // Sparse keyspace might be visited multiple times because
+                    // we don't track unmapped keyspaces.
+                    return OnDiskValueIo::Unnecessary;
+                } else {
+                    unreachable!()
                }
            }
+            ValueReconstructSituation::Continue => {
+                self.num_active_ios
+                    .fetch_add(1, std::sync::atomic::Ordering::Release);
+                let (tx, rx) = tokio::sync::oneshot::channel();
+                state.on_disk_values.push((lsn, OnDiskValueIoWaiter { rx }));
+                OnDiskValueIo::Required {
+                    tx,
+                    num_active_ios: Arc::clone(&self.num_active_ios),
+                }
+            }
+        };

-            state.situation
-        } else {
-            ValueReconstructSituation::Complete
+        if completes && state.situation == ValueReconstructSituation::Continue {
+            state.situation = ValueReconstructSituation::Complete;
+            if !is_sparse_key {
+                self.keys_done.add_key(*key);
+            }
        }
-    }

-    /// Returns the Lsn at which this key is cached if one exists.
-    /// The read path should go no further than this Lsn for the given key.
-    pub(crate) fn get_cached_lsn(&self, key: &Key) -> Option<Lsn> {
-        self.keys
-            .get(key)
-            .and_then(|k| k.as_ref().ok())
-            .and_then(|state| state.get_cached_lsn())
+        required_io
    }

    /// Returns the key space describing the keys that have
@@ -276,12 +692,6 @@ impl ValuesReconstructState {
    }
 }

-impl Default for ValuesReconstructState {
-    fn default() -> Self {
-        Self::new()
-    }
-}
-
 /// A key that uniquely identifies a layer in a timeline
 #[derive(Debug, PartialEq, Eq, Clone, Hash)]
 pub(crate) enum LayerId {
@@ -720,3 +1130,78 @@ impl<T: std::fmt::Display> std::fmt::Debug for RangeDisplayDebug<'_, T> {
        write!(f, "{}..{}", self.0.start, self.0.end)
    }
 }
+
+#[cfg(test)]
+mod tests2 {
+    use pageserver_api::key::DBDIR_KEY;
+    use tracing::info;
+
+    use super::*;
+    use crate::tenant::storage_layer::IoConcurrency;
+
+    /// TODO: currently this test relies on manual visual inspection of the --no-capture output.
+    /// Should look like so:
+    /// ```text
+    /// RUST_LOG=trace cargo nextest run  --features testing  --no-capture test_io_concurrency_noise
+    /// running 1 test
+    /// 2025-01-21T17:42:01.335679Z  INFO get_vectored_concurrent_io test selected=SidecarTask
+    /// 2025-01-21T17:42:01.335680Z TRACE spawning sidecar task task_id=0
+    /// 2025-01-21T17:42:01.335937Z TRACE IoConcurrency_sidecar{task_id=0}: start
+    /// 2025-01-21T17:42:01.335972Z TRACE IoConcurrency_sidecar{task_id=0}: received new io future
+    /// 2025-01-21T17:42:01.335999Z  INFO IoConcurrency_sidecar{task_id=0}: waiting for signal to complete IO
+    /// 2025-01-21T17:42:01.336229Z  WARN dropping ValuesReconstructState while some IOs have not been completed num_active_ios=1 sidecar_task_id=Some(0) backtrace=   0: <pageserver::tenant::storage_layer::ValuesReconstructState as core::ops::drop::Drop>::drop
+    ///              at ./src/tenant/storage_layer.rs:553:24
+    ///    1: core::ptr::drop_in_place<pageserver::tenant::storage_layer::ValuesReconstructState>
+    ///              at /home/christian/.rustup/toolchains/1.84.0-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/ptr/mod.rs:521:1
+    ///    2: core::mem::drop
+    ///              at /home/christian/.rustup/toolchains/1.84.0-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/mem/mod.rs:942:24
+    ///    3: pageserver::tenant::storage_layer::tests2::test_io_concurrency_noise::{{closure}}
+    ///              at ./src/tenant/storage_layer.rs:1159:9
+    ///   ...
+    ///   49: <unknown>
+    /// 2025-01-21T17:42:01.452293Z  INFO IoConcurrency_sidecar{task_id=0}: completing IO
+    /// 2025-01-21T17:42:01.452357Z TRACE IoConcurrency_sidecar{task_id=0}: io future completed
+    /// 2025-01-21T17:42:01.452473Z TRACE IoConcurrency_sidecar{task_id=0}: end
+    /// test tenant::storage_layer::tests2::test_io_concurrency_noise ... ok
+    ///
+    /// ```
+    #[tokio::test]
+    async fn test_io_concurrency_noise() {
+        crate::tenant::harness::setup_logging();
+
+        let io_concurrency = IoConcurrency::spawn_for_test();
+        match *io_concurrency {
+            IoConcurrency::Sequential => {
+                // This test asserts behavior in sidecar mode, doesn't make sense in sequential mode.
+                return;
+            }
+            IoConcurrency::SidecarTask { .. } => {}
+        }
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
+
+        let (io_fut_is_waiting_tx, io_fut_is_waiting) = tokio::sync::oneshot::channel();
+        let (do_complete_io, should_complete_io) = tokio::sync::oneshot::channel();
+        let (io_fut_exiting_tx, io_fut_exiting) = tokio::sync::oneshot::channel();
+
+        let io = reconstruct_state.update_key(&DBDIR_KEY, Lsn(8), true);
+        reconstruct_state
+            .spawn_io(async move {
+                info!("waiting for signal to complete IO");
+                io_fut_is_waiting_tx.send(()).unwrap();
+                should_complete_io.await.unwrap();
+                info!("completing IO");
+                io.complete(Ok(OnDiskValue::RawImage(Bytes::new())));
+                io_fut_exiting_tx.send(()).unwrap();
+            })
+            .await;
+
+        io_fut_is_waiting.await.unwrap();
+
+        // this is what makes the noise
+        drop(reconstruct_state);
+
+        do_complete_io.send(()).unwrap();
+
+        io_fut_exiting.await.unwrap();
+    }
+}
--- a/pageserver/src/tenant/storage_layer/delta_layer.rs
+++ b/pageserver/src/tenant/storage_layer/delta_layer.rs
@@ -41,13 +41,12 @@ use crate::tenant::vectored_blob_io::{
    BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
    VectoredReadPlanner,
 };
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::{FullSlice, IoBufExt};
 use crate::virtual_file::IoBufferMut;
 use crate::virtual_file::{self, MaybeFatalIo, VirtualFile};
 use crate::TEMP_FILE_SUFFIX;
 use crate::{DELTA_FILE_MAGIC, STORAGE_FORMAT_VERSION};
-use anyhow::{anyhow, bail, ensure, Context, Result};
+use anyhow::{bail, ensure, Context, Result};
 use camino::{Utf8Path, Utf8PathBuf};
 use futures::StreamExt;
 use itertools::Itertools;
@@ -60,7 +59,7 @@ use pageserver_api::shard::TenantShardId;
 use pageserver_api::value::Value;
 use rand::{distributions::Alphanumeric, Rng};
 use serde::{Deserialize, Serialize};
-use std::collections::VecDeque;
+use std::collections::{HashMap, VecDeque};
 use std::fs::File;
 use std::io::SeekFrom;
 use std::ops::Range;
@@ -77,7 +76,10 @@ use utils::{
    lsn::Lsn,
 };

-use super::{AsLayerDesc, LayerName, PersistentLayerDesc, ValuesReconstructState};
+use super::{
+    AsLayerDesc, LayerName, OnDiskValue, OnDiskValueIo, PersistentLayerDesc, ResidentLayer,
+    ValuesReconstructState,
+};

 ///
 /// Header stored in the beginning of the file
@@ -226,7 +228,7 @@ pub struct DeltaLayerInner {
    index_start_blk: u32,
    index_root_blk: u32,

-    file: VirtualFile,
+    file: Arc<VirtualFile>,
    file_id: FileId,

    layer_key_range: Range<Key>,
@@ -795,9 +797,11 @@ impl DeltaLayerInner {
        max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
        ctx: &RequestContext,
    ) -> anyhow::Result<Self> {
-        let file = VirtualFile::open_v2(path, ctx)
-            .await
-            .context("open layer file")?;
+        let file = Arc::new(
+            VirtualFile::open_v2(path, ctx)
+                .await
+                .context("open layer file")?,
+        );

        let file_id = page_cache::next_file_id();

@@ -842,12 +846,11 @@ impl DeltaLayerInner {
    // Look up the keys in the provided keyspace and update
    // the reconstruct state with whatever is found.
    //
-    // If the key is cached, go no further than the cached Lsn.
-    //
    // Currently, the index is visited for each range, but this
    // can be further optimised to visit the index only once.
    pub(super) async fn get_values_reconstruct_data(
        &self,
+        this: ResidentLayer,
        keyspace: KeySpace,
        lsn_range: Range<Lsn>,
        reconstruct_state: &mut ValuesReconstructState,
@@ -875,17 +878,14 @@ impl DeltaLayerInner {
            data_end_offset,
            index_reader,
            planner,
-            reconstruct_state,
            ctx,
        )
        .await
        .map_err(GetVectoredError::Other)?;

-        self.do_reads_and_update_state(reads, reconstruct_state, ctx)
+        self.do_reads_and_update_state(this, reads, reconstruct_state, ctx)
            .await;

-        reconstruct_state.on_lsn_advanced(&keyspace, lsn_range.start);
-
        Ok(())
    }

@@ -895,7 +895,6 @@ impl DeltaLayerInner {
        data_end_offset: u64,
        index_reader: DiskBtreeReader<Reader, DELTA_KEY_SIZE>,
        mut planner: VectoredReadPlanner,
-        reconstruct_state: &mut ValuesReconstructState,
        ctx: &RequestContext,
    ) -> anyhow::Result<Vec<VectoredRead>>
    where
@@ -922,10 +921,9 @@ impl DeltaLayerInner {
                assert!(key >= range.start);

                let outside_lsn_range = !lsn_range.contains(&lsn);
-                let below_cached_lsn = reconstruct_state.get_cached_lsn(&key) >= Some(lsn);

                let flag = {
-                    if outside_lsn_range || below_cached_lsn {
+                    if outside_lsn_range {
                        BlobFlag::Ignore
                    } else if blob_ref.will_init() {
                        BlobFlag::ReplaceAll
@@ -994,98 +992,78 @@ impl DeltaLayerInner {

    async fn do_reads_and_update_state(
        &self,
+        this: ResidentLayer,
        reads: Vec<VectoredRead>,
        reconstruct_state: &mut ValuesReconstructState,
        ctx: &RequestContext,
    ) {
-        let vectored_blob_reader = VectoredBlobReader::new(&self.file);
-        let mut ignore_key_with_err = None;
-
        let max_vectored_read_bytes = self
            .max_vectored_read_bytes
            .expect("Layer is loaded with max vectored bytes config")
            .0
            .into();
        let buf_size = Self::get_min_read_buffer_size(&reads, max_vectored_read_bytes);
-        let mut buf = Some(IoBufferMut::with_capacity(buf_size));

        // Note that reads are processed in reverse order (from highest key+lsn).
        // This is the order that `ReconstructState` requires such that it can
        // track when a key is done.
        for read in reads.into_iter().rev() {
-            let res = vectored_blob_reader
-                .read_blobs(&read, buf.take().expect("Should have a buffer"), ctx)
-                .await;
-
-            let blobs_buf = match res {
-                Ok(blobs_buf) => blobs_buf,
-                Err(err) => {
-                    let kind = err.kind();
-                    for (_, blob_meta) in read.blobs_at.as_slice() {
-                        reconstruct_state.on_key_error(
-                            blob_meta.key,
-                            PageReconstructError::Other(anyhow!(
-                                "Failed to read blobs from virtual file {}: {}",
-                                self.file.path(),
-                                kind
-                            )),
-                        );
-                    }
-
-                    // We have "lost" the buffer since the lower level IO api
-                    // doesn't return the buffer on error. Allocate a new one.
-                    buf = Some(IoBufferMut::with_capacity(buf_size));
-
-                    continue;
-                }
-            };
-            let view = BufView::new_slice(&blobs_buf.buf);
-            for meta in blobs_buf.blobs.iter().rev() {
-                if Some(meta.meta.key) == ignore_key_with_err {
-                    continue;
-                }
-                let blob_read = meta.read(&view).await;
-                let blob_read = match blob_read {
-                    Ok(buf) => buf,
-                    Err(e) => {
-                        reconstruct_state.on_key_error(
-                            meta.meta.key,
-                            PageReconstructError::Other(anyhow!(e).context(format!(
-                                "Failed to decompress blob from virtual file {}",
-                                self.file.path(),
-                            ))),
-                        );
-
-                        ignore_key_with_err = Some(meta.meta.key);
-                        continue;
-                    }
-                };
-
-                let value = Value::des(&blob_read);
-
-                let value = match value {
-                    Ok(v) => v,
-                    Err(e) => {
-                        reconstruct_state.on_key_error(
-                            meta.meta.key,
-                            PageReconstructError::Other(anyhow!(e).context(format!(
-                                "Failed to deserialize blob from virtual file {}",
-                                self.file.path(),
-                            ))),
-                        );
-
-                        ignore_key_with_err = Some(meta.meta.key);
-                        continue;
-                    }
-                };
-
-                // Invariant: once a key reaches [`ValueReconstructSituation::Complete`]
-                // state, no further updates shall be made to it. The call below will
-                // panic if the invariant is violated.
-                reconstruct_state.update_key(&meta.meta.key, meta.meta.lsn, value);
+            let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+            for (_, blob_meta) in read.blobs_at.as_slice().iter().rev() {
+                let io = reconstruct_state.update_key(
+                    &blob_meta.key,
+                    blob_meta.lsn,
+                    blob_meta.will_init,
+                );
+                ios.insert((blob_meta.key, blob_meta.lsn), io);
            }

-            buf = Some(blobs_buf.buf);
+            let read_extend_residency = this.clone();
+            let read_from = self.file.clone();
+            let read_ctx = ctx.attached_child();
+            reconstruct_state
+                .spawn_io(async move {
+                    let vectored_blob_reader = VectoredBlobReader::new(&read_from);
+                    let buf = IoBufferMut::with_capacity(buf_size);
+
+                    let res = vectored_blob_reader.read_blobs(&read, buf, &read_ctx).await;
+                    match res {
+                        Ok(blobs_buf) => {
+                            let view = BufView::new_slice(&blobs_buf.buf);
+                            for meta in blobs_buf.blobs.iter().rev() {
+                                let io = ios.remove(&(meta.meta.key, meta.meta.lsn)).unwrap();
+
+                                let blob_read = meta.read(&view).await;
+                                let blob_read = match blob_read {
+                                    Ok(buf) => buf,
+                                    Err(e) => {
+                                        io.complete(Err(e));
+                                        continue;
+                                    }
+                                };
+
+                                io.complete(Ok(OnDiskValue::WalRecordOrImage(
+                                    blob_read.into_bytes(),
+                                )));
+                            }
+
+                            assert!(ios.is_empty());
+                        }
+                        Err(err) => {
+                            for (_, sender) in ios {
+                                sender.complete(Err(std::io::Error::new(
+                                    err.kind(),
+                                    "vec read failed",
+                                )));
+                            }
+                        }
+                    }
+
+                    // keep layer resident until this IO is done; this spawned IO future generally outlives the
+                    // call to `self` / the `Arc<DownloadedLayer>` / the `ResidentLayer` that guarantees residency
+                    drop(read_extend_residency);
+                })
+                .await;
        }
    }

@@ -1224,7 +1202,14 @@ impl DeltaLayerInner {
            let actionable = if let Some((key, lsn, start_offset)) = prev.take() {
                let end_offset = offset;

-                Some((BlobMeta { key, lsn }, start_offset..end_offset))
+                Some((
+                    BlobMeta {
+                        key,
+                        lsn,
+                        will_init: false,
+                    },
+                    start_offset..end_offset,
+                ))
            } else {
                None
            };
@@ -1560,7 +1545,9 @@ impl DeltaLayerIterator<'_> {
                let lsn = DeltaKey::extract_lsn_from_buf(&raw_key);
                let blob_ref = BlobRef(value);
                let offset = blob_ref.pos();
-                if let Some(batch_plan) = self.planner.handle(key, lsn, offset) {
+                if let Some(batch_plan) =
+                    self.planner.handle(key, lsn, offset, blob_ref.will_init())
+                {
                    break batch_plan;
                }
            } else {
@@ -1673,7 +1660,6 @@ pub(crate) mod test {
            .expect("In memory disk finish should never fail");
        let reader = DiskBtreeReader::<_, DELTA_KEY_SIZE>::new(0, root_offset, disk);
        let planner = VectoredReadPlanner::new(100);
-        let mut reconstruct_state = ValuesReconstructState::new();
        let ctx = RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Error);

        let keyspace = KeySpace {
@@ -1691,7 +1677,6 @@ pub(crate) mod test {
            disk_offset,
            reader,
            planner,
-            &mut reconstruct_state,
            &ctx,
        )
        .await
@@ -1935,7 +1920,6 @@ pub(crate) mod test {
            );

            let planner = VectoredReadPlanner::new(constants::MAX_VECTORED_READ_BYTES);
-            let mut reconstruct_state = ValuesReconstructState::new();
            let keyspace = pick_random_keyspace(rng, &entries_meta.key_range);
            let data_end_offset = inner.index_start_blk as u64 * PAGE_SZ as u64;

@@ -1945,7 +1929,6 @@ pub(crate) mod test {
                data_end_offset,
                index_reader,
                planner,
-                &mut reconstruct_state,
                &ctx,
            )
            .await?;
--- a/pageserver/src/tenant/storage_layer/image_layer.rs
+++ b/pageserver/src/tenant/storage_layer/image_layer.rs
@@ -38,12 +38,11 @@ use crate::tenant::vectored_blob_io::{
    BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
    VectoredReadPlanner,
 };
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
 use crate::virtual_file::IoBufferMut;
 use crate::virtual_file::{self, MaybeFatalIo, VirtualFile};
 use crate::{IMAGE_FILE_MAGIC, STORAGE_FORMAT_VERSION, TEMP_FILE_SUFFIX};
-use anyhow::{anyhow, bail, ensure, Context, Result};
+use anyhow::{bail, ensure, Context, Result};
 use bytes::Bytes;
 use camino::{Utf8Path, Utf8PathBuf};
 use hex;
@@ -56,12 +55,13 @@ use pageserver_api::shard::{ShardIdentity, TenantShardId};
 use pageserver_api::value::Value;
 use rand::{distributions::Alphanumeric, Rng};
 use serde::{Deserialize, Serialize};
-use std::collections::VecDeque;
+use std::collections::{HashMap, VecDeque};
 use std::fs::File;
 use std::io::SeekFrom;
 use std::ops::Range;
 use std::os::unix::prelude::FileExt;
 use std::str::FromStr;
+use std::sync::Arc;
 use tokio::sync::OnceCell;
 use tokio_stream::StreamExt;
 use tracing::*;
@@ -73,7 +73,10 @@ use utils::{
 };

 use super::layer_name::ImageLayerName;
-use super::{AsLayerDesc, LayerName, PersistentLayerDesc, ValuesReconstructState};
+use super::{
+    AsLayerDesc, LayerName, OnDiskValue, OnDiskValueIo, PersistentLayerDesc, ResidentLayer,
+    ValuesReconstructState,
+};

 ///
 /// Header stored in the beginning of the file
@@ -164,7 +167,7 @@ pub struct ImageLayerInner {
    key_range: Range<Key>,
    lsn: Lsn,

-    file: VirtualFile,
+    file: Arc<VirtualFile>,
    file_id: FileId,

    max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
@@ -391,9 +394,11 @@ impl ImageLayerInner {
        max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
        ctx: &RequestContext,
    ) -> anyhow::Result<Self> {
-        let file = VirtualFile::open_v2(path, ctx)
-            .await
-            .context("open layer file")?;
+        let file = Arc::new(
+            VirtualFile::open_v2(path, ctx)
+                .await
+                .context("open layer file")?,
+        );
        let file_id = page_cache::next_file_id();
        let block_reader = FileBlockReader::new(&file, file_id);
        let summary_blk = block_reader
@@ -439,6 +444,7 @@ impl ImageLayerInner {
    // the reconstruct state with whatever is found.
    pub(super) async fn get_values_reconstruct_data(
        &self,
+        this: ResidentLayer,
        keyspace: KeySpace,
        reconstruct_state: &mut ValuesReconstructState,
        ctx: &RequestContext,
@@ -448,7 +454,7 @@ impl ImageLayerInner {
            .await
            .map_err(GetVectoredError::Other)?;

-        self.do_reads_and_update_state(reads, reconstruct_state, ctx)
+        self.do_reads_and_update_state(this, reads, reconstruct_state, ctx)
            .await;

        reconstruct_state.on_image_layer_visited(&self.key_range);
@@ -570,6 +576,7 @@ impl ImageLayerInner {

    async fn do_reads_and_update_state(
        &self,
+        this: ResidentLayer,
        reads: Vec<VectoredRead>,
        reconstruct_state: &mut ValuesReconstructState,
        ctx: &RequestContext,
@@ -580,8 +587,13 @@ impl ImageLayerInner {
            .0
            .into();

-        let vectored_blob_reader = VectoredBlobReader::new(&self.file);
        for read in reads.into_iter() {
+            let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+            for (_, blob_meta) in read.blobs_at.as_slice() {
+                let io = reconstruct_state.update_key(&blob_meta.key, blob_meta.lsn, true);
+                ios.insert((blob_meta.key, blob_meta.lsn), io);
+            }
+
            let buf_size = read.size();

            if buf_size > max_vectored_read_bytes {
@@ -611,50 +623,51 @@ impl ImageLayerInner {
                }
            }

-            let buf = IoBufferMut::with_capacity(buf_size);
-            let res = vectored_blob_reader.read_blobs(&read, buf, ctx).await;
+            let read_extend_residency = this.clone();
+            let read_from = self.file.clone();
+            let read_ctx = ctx.attached_child();
+            reconstruct_state
+                .spawn_io(async move {
+                    let buf = IoBufferMut::with_capacity(buf_size);
+                    let vectored_blob_reader = VectoredBlobReader::new(&read_from);
+                    let res = vectored_blob_reader.read_blobs(&read, buf, &read_ctx).await;

-            match res {
-                Ok(blobs_buf) => {
-                    let view = BufView::new_slice(&blobs_buf.buf);
-                    for meta in blobs_buf.blobs.iter() {
-                        let img_buf = meta.read(&view).await;
+                    match res {
+                        Ok(blobs_buf) => {
+                            let view = BufView::new_slice(&blobs_buf.buf);
+                            for meta in blobs_buf.blobs.iter() {
+                                let io: OnDiskValueIo =
+                                    ios.remove(&(meta.meta.key, meta.meta.lsn)).unwrap();
+                                let img_buf = meta.read(&view).await;

-                        let img_buf = match img_buf {
-                            Ok(img_buf) => img_buf,
-                            Err(e) => {
-                                reconstruct_state.on_key_error(
-                                    meta.meta.key,
-                                    PageReconstructError::Other(anyhow!(e).context(format!(
-                                        "Failed to decompress blob from virtual file {}",
-                                        self.file.path(),
-                                    ))),
-                                );
+                                let img_buf = match img_buf {
+                                    Ok(img_buf) => img_buf,
+                                    Err(e) => {
+                                        io.complete(Err(e));
+                                        continue;
+                                    }
+                                };

-                                continue;
+                                io.complete(Ok(OnDiskValue::RawImage(img_buf.into_bytes())));
                            }
-                        };
-                        reconstruct_state.update_key(
-                            &meta.meta.key,
-                            self.lsn,
-                            Value::Image(img_buf.into_bytes()),
-                        );
+
+                            assert!(ios.is_empty());
+                        }
+                        Err(err) => {
+                            for (_, io) in ios {
+                                io.complete(Err(std::io::Error::new(
+                                    err.kind(),
+                                    "vec read failed",
+                                )));
+                            }
+                        }
                    }
-                }
-                Err(err) => {
-                    let kind = err.kind();
-                    for (_, blob_meta) in read.blobs_at.as_slice() {
-                        reconstruct_state.on_key_error(
-                            blob_meta.key,
-                            PageReconstructError::from(anyhow!(
-                                "Failed to read blobs from virtual file {}: {}",
-                                self.file.path(),
-                                kind
-                            )),
-                        );
-                    }
-                }
-            };
+
+                    // keep layer resident until this IO is done; this spawned IO future generally outlives the
+                    // call to `self` / the `Arc<DownloadedLayer>` / the `ResidentLayer` that guarantees residency
+                    drop(read_extend_residency);
+                })
+                .await;
        }
    }

@@ -1069,6 +1082,7 @@ impl ImageLayerIterator<'_> {
                    Key::from_slice(&raw_key[..KEY_SIZE]),
                    self.image_layer.lsn,
                    offset,
+                    true,
                ) {
                    break batch_plan;
                }
--- a/pageserver/src/tenant/storage_layer/inmemory_layer.rs
+++ b/pageserver/src/tenant/storage_layer/inmemory_layer.rs
@@ -8,23 +8,22 @@ use crate::assert_u64_eq_usize::{u64_to_usize, U64IsUsize, UsizeIsU64};
 use crate::config::PageServerConf;
 use crate::context::{PageContentKind, RequestContext, RequestContextBuilder};
 use crate::tenant::ephemeral_file::EphemeralFile;
+use crate::tenant::storage_layer::{OnDiskValue, OnDiskValueIo};
 use crate::tenant::timeline::GetVectoredError;
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
 use crate::{l0_flush, page_cache};
-use anyhow::{anyhow, Result};
+use anyhow::Result;
 use camino::Utf8PathBuf;
 use pageserver_api::key::CompactKey;
 use pageserver_api::key::Key;
 use pageserver_api::keyspace::KeySpace;
 use pageserver_api::models::InMemoryLayerInfo;
 use pageserver_api::shard::TenantShardId;
-use pageserver_api::value::Value;
 use std::collections::{BTreeMap, HashMap};
 use std::sync::{Arc, OnceLock};
 use std::time::Instant;
 use tracing::*;
-use utils::{bin_ser::BeSer, id::TimelineId, lsn::Lsn, vec_map::VecMap};
+use utils::{id::TimelineId, lsn::Lsn, vec_map::VecMap};
 use wal_decoder::serialized_batch::{SerializedValueBatch, SerializedValueMeta, ValueMeta};
 // avoid binding to Write (conflicts with std::io::Write)
 // while being able to use std::fmt::Write's methods
@@ -36,9 +35,7 @@ use std::sync::atomic::Ordering as AtomicOrdering;
 use std::sync::atomic::{AtomicU64, AtomicUsize};
 use tokio::sync::RwLock;

-use super::{
-    DeltaLayerWriter, PersistentLayerDesc, ValueReconstructSituation, ValuesReconstructState,
-};
+use super::{DeltaLayerWriter, PersistentLayerDesc, ValuesReconstructState};

 pub(crate) mod vectored_dio_read;

@@ -415,10 +412,8 @@ impl InMemoryLayer {

    // Look up the keys in the provided keyspace and update
    // the reconstruct state with whatever is found.
-    //
-    // If the key is cached, go no further than the cached Lsn.
    pub(crate) async fn get_values_reconstruct_data(
-        &self,
+        self: &Arc<InMemoryLayer>,
        keyspace: KeySpace,
        end_lsn: Lsn,
        reconstruct_state: &mut ValuesReconstructState,
@@ -435,6 +430,9 @@ impl InMemoryLayer {
            read: vectored_dio_read::LogicalRead<Vec<u8>>,
        }
        let mut reads: HashMap<Key, Vec<ValueRead>> = HashMap::new();
+        let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+
+        let lsn_range = self.start_lsn..end_lsn;

        for range in keyspace.ranges.iter() {
            for (key, vec_map) in inner
@@ -442,12 +440,7 @@ impl InMemoryLayer {
                .range(range.start.to_compact()..range.end.to_compact())
            {
                let key = Key::from_compact(*key);
-                let lsn_range = match reconstruct_state.get_cached_lsn(&key) {
-                    Some(cached_lsn) => (cached_lsn + 1)..end_lsn,
-                    None => self.start_lsn..end_lsn,
-                };
-
-                let slice = vec_map.slice_range(lsn_range);
+                let slice = vec_map.slice_range(lsn_range.clone());

                for (entry_lsn, index_entry) in slice.iter().rev() {
                    let IndexEntryUnpacked {
@@ -463,55 +456,59 @@ impl InMemoryLayer {
                            Vec::with_capacity(len as usize),
                        ),
                    });
+
+                    let io = reconstruct_state.update_key(&key, *entry_lsn, will_init);
+                    ios.insert((key, *entry_lsn), io);
+
                    if will_init {
                        break;
                    }
                }
            }
        }
+        drop(inner); // release the lock before we spawn the IO; if it's serial-mode IO we will deadlock on the read().await below
+        let read_from = Arc::clone(self);
+        let read_ctx = ctx.attached_child();
+        reconstruct_state
+            .spawn_io(async move {
+                let inner = read_from.inner.read().await;
+                let f = vectored_dio_read::execute(
+                    &inner.file,
+                    reads
+                        .iter()
+                        .flat_map(|(_, value_reads)| value_reads.iter().map(|v| &v.read)),
+                    &read_ctx,
+                );
+                send_future::SendFuture::send(f) // https://github.com/rust-lang/rust/issues/96865
+                    .await;

-        // Execute the reads.
-
-        let f = vectored_dio_read::execute(
-            &inner.file,
-            reads
-                .iter()
-                .flat_map(|(_, value_reads)| value_reads.iter().map(|v| &v.read)),
-            &ctx,
-        );
-        send_future::SendFuture::send(f) // https://github.com/rust-lang/rust/issues/96865
-            .await;
-
-        // Process results into the reconstruct state
-        'next_key: for (key, value_reads) in reads {
-            for ValueRead { entry_lsn, read } in value_reads {
-                match read.into_result().expect("we run execute() above") {
-                    Err(e) => {
-                        reconstruct_state.on_key_error(key, PageReconstructError::from(anyhow!(e)));
-                        continue 'next_key;
-                    }
-                    Ok(value_buf) => {
-                        let value = Value::des(&value_buf);
-                        if let Err(e) = value {
-                            reconstruct_state
-                                .on_key_error(key, PageReconstructError::from(anyhow!(e)));
-                            continue 'next_key;
+                for (key, value_reads) in reads {
+                    for ValueRead { entry_lsn, read } in value_reads {
+                        let io = ios.remove(&(key, entry_lsn)).expect("sender must exist");
+                        match read.into_result().expect("we run execute() above") {
+                            Err(e) => {
+                                io.complete(Err(std::io::Error::new(
+                                    e.kind(),
+                                    "dio vec read failed",
+                                )));
+                            }
+                            Ok(value_buf) => {
+                                io.complete(Ok(OnDiskValue::WalRecordOrImage(value_buf.into())));
+                            }
                        }
-
-                        let key_situation =
-                            reconstruct_state.update_key(&key, entry_lsn, value.unwrap());
-                        if key_situation == ValueReconstructSituation::Complete {
-                            // TODO: metric to see if we fetched more values than necessary
-                            continue 'next_key;
-                        }
-
-                        // process the next value in the next iteration of the loop
                    }
                }
-            }
-        }

-        reconstruct_state.on_lsn_advanced(&keyspace, self.start_lsn);
+                assert!(ios.is_empty());
+
+                // Keep layer existent until this IO is done;
+                // This is kinda forced for InMemoryLayer because we need to inner.read() anyway,
+                // but it's less obvious for DeltaLayer and ImageLayer. So, keep this explicit
+                // drop for consistency among all three layer types.
+                drop(inner);
+                drop(read_from);
+            })
+            .await;

        Ok(())
    }
@@ -606,6 +603,7 @@ impl InMemoryLayer {
        // Write the batch to the file
        inner.file.write_raw(&raw, ctx).await?;
        let new_size = inner.file.len();
+
        let expected_new_len = base_offset
            .checked_add(raw.len().into_u64())
            // write_raw would error if we were to overflow u64.
--- a/pageserver/src/tenant/storage_layer/layer.rs
+++ b/pageserver/src/tenant/storage_layer/layer.rs
@@ -308,7 +308,7 @@ impl Layer {
        reconstruct_data: &mut ValuesReconstructState,
        ctx: &RequestContext,
    ) -> Result<(), GetVectoredError> {
-        let layer = self
+        let downloaded = self
            .0
            .get_or_maybe_download(true, Some(ctx))
            .await
@@ -318,11 +318,15 @@ impl Layer {
                }
                other => GetVectoredError::Other(anyhow::anyhow!(other)),
            })?;
+        let this = ResidentLayer {
+            downloaded: downloaded.clone(),
+            owner: self.clone(),
+        };

        self.record_access(ctx);

-        layer
-            .get_values_reconstruct_data(keyspace, lsn_range, reconstruct_data, &self.0, ctx)
+        downloaded
+            .get_values_reconstruct_data(this, keyspace, lsn_range, reconstruct_data, ctx)
            .instrument(tracing::debug_span!("get_values_reconstruct_data", layer=%self))
            .await
            .map_err(|err| match err {
@@ -1768,25 +1772,25 @@ impl DownloadedLayer {

    async fn get_values_reconstruct_data(
        &self,
+        this: ResidentLayer,
        keyspace: KeySpace,
        lsn_range: Range<Lsn>,
        reconstruct_data: &mut ValuesReconstructState,
-        owner: &Arc<LayerInner>,
        ctx: &RequestContext,
    ) -> Result<(), GetVectoredError> {
        use LayerKind::*;

        match self
-            .get(owner, ctx)
+            .get(&this.owner.0, ctx)
            .await
            .map_err(GetVectoredError::Other)?
        {
            Delta(d) => {
-                d.get_values_reconstruct_data(keyspace, lsn_range, reconstruct_data, ctx)
+                d.get_values_reconstruct_data(this, keyspace, lsn_range, reconstruct_data, ctx)
                    .await
            }
            Image(i) => {
-                i.get_values_reconstruct_data(keyspace, reconstruct_data, ctx)
+                i.get_values_reconstruct_data(this, keyspace, reconstruct_data, ctx)
                    .await
            }
        }
--- a/pageserver/src/tenant/storage_layer/layer/tests.rs
+++ b/pageserver/src/tenant/storage_layer/layer/tests.rs
@@ -11,7 +11,10 @@ use super::failpoints::{Failpoint, FailpointKind};
 use super::*;
 use crate::{
    context::DownloadBehavior,
-    tenant::{harness::test_img, storage_layer::LayerVisibilityHint},
+    tenant::{
+        harness::test_img,
+        storage_layer::{IoConcurrency, LayerVisibilityHint},
+    },
 };
 use crate::{task_mgr::TaskKind, tenant::harness::TenantHarness};

@@ -31,6 +34,7 @@ async fn smoke_test() {
    let span = h.span();
    let download_span = span.in_scope(|| tracing::info_span!("downloading", timeline_id = 1));
    let (tenant, _) = h.load().await;
+    let io_concurrency = IoConcurrency::spawn_for_test();

    let ctx = RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Download);

@@ -89,7 +93,7 @@ async fn smoke_test() {
    };

    let img_before = {
-        let mut data = ValuesReconstructState::default();
+        let mut data = ValuesReconstructState::new(io_concurrency.clone());
        layer
            .get_values_reconstruct_data(
                controlfile_keyspace.clone(),
@@ -99,10 +103,13 @@ async fn smoke_test() {
            )
            .await
            .unwrap();
+
        data.keys
            .remove(&CONTROLFILE_KEY)
            .expect("must be present")
-            .expect("should not error")
+            .collect_pending_ios()
+            .await
+            .expect("must not error")
            .img
            .take()
            .expect("tenant harness writes the control file")
@@ -121,7 +128,7 @@ async fn smoke_test() {

    // on accesses when the layer is evicted, it will automatically be downloaded.
    let img_after = {
-        let mut data = ValuesReconstructState::default();
+        let mut data = ValuesReconstructState::new(io_concurrency.clone());
        layer
            .get_values_reconstruct_data(
                controlfile_keyspace.clone(),
@@ -135,7 +142,9 @@ async fn smoke_test() {
        data.keys
            .remove(&CONTROLFILE_KEY)
            .expect("must be present")
-            .expect("should not error")
+            .collect_pending_ios()
+            .await
+            .expect("must not error")
            .img
            .take()
            .expect("tenant harness writes the control file")
--- a/pageserver/src/tenant/tasks.rs
+++ b/pageserver/src/tenant/tasks.rs
@@ -67,10 +67,9 @@ pub(crate) async fn concurrent_background_tasks_rate_limit_permit(
 ) -> tokio::sync::SemaphorePermit<'static> {
    let _guard = crate::metrics::BACKGROUND_LOOP_SEMAPHORE.measure_acquisition(loop_kind);

-    pausable_failpoint!(
-        "initial-size-calculation-permit-pause",
-        loop_kind == BackgroundLoopKind::InitialLogicalSizeCalculation
-    );
+    if loop_kind == BackgroundLoopKind::InitialLogicalSizeCalculation {
+        pausable_failpoint!("initial-size-calculation-permit-pause");
+    }

    // TODO: assert that we run on BACKGROUND_RUNTIME; requires tokio_unstable Handle::id();
    match CONCURRENT_BACKGROUND_TASKS.acquire().await {
--- a/pageserver/src/tenant/timeline.rs
+++ b/pageserver/src/tenant/timeline.rs
@@ -20,12 +20,13 @@ use camino::Utf8Path;
 use chrono::{DateTime, Utc};
 use enumset::EnumSet;
 use fail::fail_point;
+use futures::{stream::FuturesUnordered, StreamExt};
 use handle::ShardTimelineId;
+use layer_manager::Shutdown;
 use offload::OffloadError;
 use once_cell::sync::Lazy;
 use pageserver_api::models::PageTraceEvent;
 use pageserver_api::{
-    config::tenant_conf_defaults::DEFAULT_COMPACTION_THRESHOLD,
    key::{
        KEY_SIZE, METADATA_KEY_BEGIN_PREFIX, METADATA_KEY_END_PREFIX, NON_INHERITED_RANGE,
        SPARSE_RANGE,
@@ -59,21 +60,16 @@ use utils::{
 };
 use wal_decoder::serialized_batch::{SerializedValueBatch, ValueMeta};

-use std::sync::atomic::Ordering as AtomicOrdering;
-use std::sync::{Arc, Mutex, RwLock, Weak};
+use std::array;
+use std::cmp::{max, min};
+use std::collections::btree_map::Entry;
+use std::collections::{BTreeMap, HashMap, HashSet};
+use std::ops::{ControlFlow, Deref, Range};
+use std::sync::atomic::{AtomicBool, AtomicU64, Ordering as AtomicOrdering};
+use std::sync::{Arc, Mutex, OnceLock, RwLock, Weak};
 use std::time::{Duration, Instant, SystemTime};
-use std::{
-    array,
-    collections::{BTreeMap, HashMap, HashSet},
-    sync::atomic::AtomicU64,
-};
-use std::{cmp::min, ops::ControlFlow};
-use std::{
-    collections::btree_map::Entry,
-    ops::{Deref, Range},
-};
-use std::{pin::pin, sync::OnceLock};

+use crate::l0_flush::{self, L0FlushGlobalState};
 use crate::{
    aux_file::AuxFileSizeEstimator,
    page_service::TenantManagerTypes,
@@ -81,7 +77,10 @@ use crate::{
        config::AttachmentMode,
        layer_map::{LayerMap, SearchResult},
        metadata::TimelineMetadata,
-        storage_layer::{inmemory_layer::IndexEntry, PersistentLayerDesc},
+        storage_layer::{
+            inmemory_layer::IndexEntry, IoConcurrency, PersistentLayerDesc,
+            ValueReconstructSituation,
+        },
    },
    walingest::WalLagCooldown,
    walredo,
@@ -102,10 +101,6 @@ use crate::{
 use crate::{
    disk_usage_eviction_task::EvictionCandidate, tenant::storage_layer::delta_layer::DeltaEntry,
 };
-use crate::{
-    l0_flush::{self, L0FlushGlobalState},
-    metrics::GetKind,
-};
 use crate::{
    metrics::ScanLatencyOngoingRecording, tenant::timeline::logical_size::CurrentLogicalSize,
 };
@@ -407,6 +402,9 @@ pub struct Timeline {
    /// Timeline deletion will acquire both compaction and gc locks in whatever order.
    compaction_lock: tokio::sync::Mutex<()>,

+    /// If true, the last compaction failed.
+    compaction_failed: AtomicBool,
+
    /// Make sure we only have one running gc at a time.
    ///
    /// Must only be taken in two places:
@@ -903,10 +901,17 @@ impl From<GetReadyAncestorError> for PageReconstructError {
    }
 }

+pub(crate) enum WaitLsnTimeout {
+    Custom(Duration),
+    // Use the [`PageServerConf::wait_lsn_timeout`] default
+    Default,
+}
+
 pub(crate) enum WaitLsnWaiter<'a> {
    Timeline(&'a Timeline),
    Tenant,
    PageService,
+    HttpEndpoint,
 }

 /// Argument to [`Timeline::shutdown`].
@@ -1005,9 +1010,7 @@ impl Timeline {
            ranges: vec![key..key.next()],
        };

-        // Initialise the reconstruct state for the key with the cache
-        // entry returned above.
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let mut reconstruct_state = ValuesReconstructState::new(IoConcurrency::sequential());

        let vectored_res = self
            .get_vectored_impl(keyspace.clone(), lsn, &mut reconstruct_state, ctx)
@@ -1050,6 +1053,7 @@ impl Timeline {
        &self,
        keyspace: KeySpace,
        lsn: Lsn,
+        io_concurrency: super::storage_layer::IoConcurrency,
        ctx: &RequestContext,
    ) -> Result<BTreeMap<Key, Result<Bytes, PageReconstructError>>, GetVectoredError> {
        if !lsn.is_valid() {
@@ -1084,7 +1088,7 @@ impl Timeline {
            .get_vectored_impl(
                keyspace.clone(),
                lsn,
-                &mut ValuesReconstructState::new(),
+                &mut ValuesReconstructState::new(io_concurrency),
                ctx,
            )
            .await;
@@ -1109,6 +1113,7 @@ impl Timeline {
        keyspace: KeySpace,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: super::storage_layer::IoConcurrency,
    ) -> Result<BTreeMap<Key, Result<Bytes, PageReconstructError>>, GetVectoredError> {
        if !lsn.is_valid() {
            return Err(GetVectoredError::InvalidLsn(lsn));
@@ -1140,7 +1145,7 @@ impl Timeline {
            .get_vectored_impl(
                keyspace.clone(),
                lsn,
-                &mut ValuesReconstructState::default(),
+                &mut ValuesReconstructState::new(io_concurrency),
                ctx,
            )
            .await;
@@ -1152,46 +1157,63 @@ impl Timeline {
        vectored_res
    }

-    pub(super) async fn get_vectored_impl(
+    pub(crate) async fn get_vectored_impl(
        &self,
        keyspace: KeySpace,
        lsn: Lsn,
        reconstruct_state: &mut ValuesReconstructState,
        ctx: &RequestContext,
    ) -> Result<BTreeMap<Key, Result<Bytes, PageReconstructError>>, GetVectoredError> {
-        let get_kind = if keyspace.total_raw_size() == 1 {
-            GetKind::Singular
-        } else {
-            GetKind::Vectored
+        let traversal_res: Result<(), _> = self
+            .get_vectored_reconstruct_data(keyspace.clone(), lsn, reconstruct_state, ctx)
+            .await;
+        if let Err(err) = traversal_res {
+            // Wait for all the spawned IOs to complete.
+            // See comments on `spawn_io` inside `storage_layer` for more details.
+            let mut collect_futs = std::mem::take(&mut reconstruct_state.keys)
+                .into_values()
+                .map(|state| state.collect_pending_ios())
+                .collect::<FuturesUnordered<_>>();
+            while collect_futs.next().await.is_some() {}
+            return Err(err);
        };

-        let get_data_timer = crate::metrics::GET_RECONSTRUCT_DATA_TIME
-            .for_get_kind(get_kind)
-            .start_timer();
-        self.get_vectored_reconstruct_data(keyspace.clone(), lsn, reconstruct_state, ctx)
-            .await?;
-        get_data_timer.stop_and_record();
-
-        let reconstruct_timer = crate::metrics::RECONSTRUCT_TIME
-            .for_get_kind(get_kind)
-            .start_timer();
-        let mut results: BTreeMap<Key, Result<Bytes, PageReconstructError>> = BTreeMap::new();
        let layers_visited = reconstruct_state.get_layers_visited();

-        for (key, res) in std::mem::take(&mut reconstruct_state.keys) {
-            match res {
-                Err(err) => {
-                    results.insert(key, Err(err));
-                }
-                Ok(state) => {
-                    let state = ValueReconstructState::from(state);
+        let futs = FuturesUnordered::new();
+        for (key, state) in std::mem::take(&mut reconstruct_state.keys) {
+            futs.push({
+                let walredo_self = self.myself.upgrade().expect("&self method holds the arc");
+                async move {
+                    assert_eq!(state.situation, ValueReconstructSituation::Complete);

-                    let reconstruct_res = self.reconstruct_value(key, lsn, state).await;
-                    results.insert(key, reconstruct_res);
+                    let converted = match state.collect_pending_ios().await {
+                        Ok(ok) => ok,
+                        Err(err) => {
+                            return (key, Err(err));
+                        }
+                    };
+
+                    // The walredo module expects the records to be descending in terms of Lsn.
+                    // And we submit the IOs in that order, so, there shuold be no need to sort here.
+                    debug_assert!(
+                        converted
+                            .records
+                            .is_sorted_by_key(|(lsn, _)| std::cmp::Reverse(*lsn)),
+                        "{converted:?}"
+                    );
+
+                    (
+                        key,
+                        walredo_self.reconstruct_value(key, lsn, converted).await,
+                    )
                }
-            }
+            });
        }
-        reconstruct_timer.stop_and_record();
+
+        let results = futs
+            .collect::<BTreeMap<Key, Result<Bytes, PageReconstructError>>>()
+            .await;

        // For aux file keys (v1 or v2) the vectored read path does not return an error
        // when they're missing. Instead they are omitted from the resulting btree
@@ -1286,6 +1308,7 @@ impl Timeline {
        &self,
        lsn: Lsn,
        who_is_waiting: WaitLsnWaiter<'_>,
+        timeout: WaitLsnTimeout,
        ctx: &RequestContext, /* Prepare for use by cancellation */
    ) -> Result<(), WaitLsnError> {
        let state = self.current_state();
@@ -1302,7 +1325,7 @@ impl Timeline {
                | TaskKind::WalReceiverConnectionPoller => {
                    let is_myself = match who_is_waiting {
                        WaitLsnWaiter::Timeline(waiter) => Weak::ptr_eq(&waiter.myself, &self.myself),
-                        WaitLsnWaiter::Tenant | WaitLsnWaiter::PageService => unreachable!("tenant or page_service context are not expected to have task kind {:?}", ctx.task_kind()),
+                        WaitLsnWaiter::Tenant | WaitLsnWaiter::PageService | WaitLsnWaiter::HttpEndpoint => unreachable!("tenant or page_service context are not expected to have task kind {:?}", ctx.task_kind()),
                    };
                    if is_myself {
                        if let Err(current) = self.last_record_lsn.would_wait_for(lsn) {
@@ -1318,13 +1341,14 @@ impl Timeline {
            }
        }

+        let timeout = match timeout {
+            WaitLsnTimeout::Custom(t) => t,
+            WaitLsnTimeout::Default => self.conf.wait_lsn_timeout,
+        };
+
        let _timer = crate::metrics::WAIT_LSN_TIME.start_timer();

-        match self
-            .last_record_lsn
-            .wait_for_timeout(lsn, self.conf.wait_lsn_timeout)
-            .await
-        {
+        match self.last_record_lsn.wait_for_timeout(lsn, timeout).await {
            Ok(()) => Ok(()),
            Err(e) => {
                use utils::seqwait::SeqWaitError::*;
@@ -1684,13 +1708,27 @@ impl Timeline {
            return Ok(false);
        }

-        match self.get_compaction_algorithm_settings().kind {
+        let result = match self.get_compaction_algorithm_settings().kind {
            CompactionAlgorithm::Tiered => {
                self.compact_tiered(cancel, ctx).await?;
                Ok(false)
            }
            CompactionAlgorithm::Legacy => self.compact_legacy(cancel, options, ctx).await,
-        }
+        };
+
+        // Signal compaction failure to avoid L0 flush stalls when it's broken.
+        match result {
+            Ok(_) => self.compaction_failed.store(false, AtomicOrdering::Relaxed),
+            Err(CompactionError::Other(_)) => {
+                self.compaction_failed.store(true, AtomicOrdering::Relaxed)
+            }
+            // Don't change the current value on offload failure or shutdown. We don't want to
+            // abruptly stall nor resume L0 flushes in these cases.
+            Err(CompactionError::Offload(_)) => {}
+            Err(CompactionError::ShuttingDown) => {}
+        };
+
+        result
    }

    /// Mutate the timeline with a [`TimelineWriter`].
@@ -2119,6 +2157,13 @@ impl Timeline {
            .unwrap_or(self.conf.default_tenant_conf.checkpoint_timeout)
    }

+    fn get_compaction_period(&self) -> Duration {
+        let tenant_conf = self.tenant_conf.load().tenant_conf.clone();
+        tenant_conf
+            .compaction_period
+            .unwrap_or(self.conf.default_tenant_conf.compaction_period)
+    }
+
    fn get_compaction_target_size(&self) -> u64 {
        let tenant_conf = self.tenant_conf.load();
        tenant_conf
@@ -2135,6 +2180,93 @@ impl Timeline {
            .unwrap_or(self.conf.default_tenant_conf.compaction_threshold)
    }

+    fn get_l0_flush_delay_threshold(&self) -> Option<usize> {
+        // Disable L0 flushes by default. This and compaction needs further tuning.
+        const DEFAULT_L0_FLUSH_DELAY_FACTOR: usize = 0; // TODO: default to e.g. 3
+
+        // If compaction is disabled, don't delay.
+        if self.get_compaction_period() == Duration::ZERO {
+            return None;
+        }
+
+        let compaction_threshold = self.get_compaction_threshold();
+        let tenant_conf = self.tenant_conf.load();
+        let l0_flush_delay_threshold = tenant_conf
+            .tenant_conf
+            .l0_flush_delay_threshold
+            .or(self.conf.default_tenant_conf.l0_flush_delay_threshold)
+            .unwrap_or(DEFAULT_L0_FLUSH_DELAY_FACTOR * compaction_threshold);
+
+        // 0 disables backpressure.
+        if l0_flush_delay_threshold == 0 {
+            return None;
+        }
+
+        // Clamp the flush delay threshold to the compaction threshold; it doesn't make sense to
+        // backpressure flushes below this.
+        // TODO: the tenant config should have validation to prevent this instead.
+        debug_assert!(l0_flush_delay_threshold >= compaction_threshold);
+        Some(max(l0_flush_delay_threshold, compaction_threshold))
+    }
+
+    fn get_l0_flush_stall_threshold(&self) -> Option<usize> {
+        // Disable L0 stalls by default. In ingest benchmarks, we see image compaction take >10
+        // minutes, blocking L0 compaction, and we can't stall L0 flushes for that long.
+        const DEFAULT_L0_FLUSH_STALL_FACTOR: usize = 0; // TODO: default to e.g. 5
+
+        // If compaction is disabled, don't stall.
+        if self.get_compaction_period() == Duration::ZERO {
+            return None;
+        }
+
+        // If compaction is failing, don't stall and try to keep the tenant alive. This may not be a
+        // good idea: read amp can grow unbounded, leading to terrible performance, and we may take
+        // on unbounded compaction debt that can take a long time to fix once compaction comes back
+        // online. At least we'll delay flushes, slowing down the growth and buying some time.
+        if self.compaction_failed.load(AtomicOrdering::Relaxed) {
+            return None;
+        }
+
+        let compaction_threshold = self.get_compaction_threshold();
+        let tenant_conf = self.tenant_conf.load();
+        let l0_flush_stall_threshold = tenant_conf
+            .tenant_conf
+            .l0_flush_stall_threshold
+            .or(self.conf.default_tenant_conf.l0_flush_stall_threshold);
+
+        // Tests sometimes set compaction_threshold=1 to generate lots of layer files, and don't
+        // handle the 20-second compaction delay. Some (e.g. `test_backward_compatibility`) can't
+        // easily adjust the L0 backpressure settings, so just disable stalls in this case.
+        if cfg!(feature = "testing")
+            && compaction_threshold == 1
+            && l0_flush_stall_threshold.is_none()
+        {
+            return None;
+        }
+
+        let l0_flush_stall_threshold = l0_flush_stall_threshold
+            .unwrap_or(DEFAULT_L0_FLUSH_STALL_FACTOR * compaction_threshold);
+
+        // 0 disables backpressure.
+        if l0_flush_stall_threshold == 0 {
+            return None;
+        }
+
+        // Clamp the flush stall threshold to the compaction threshold; it doesn't make sense to
+        // backpressure flushes below this.
+        // TODO: the tenant config should have validation to prevent this instead.
+        debug_assert!(l0_flush_stall_threshold >= compaction_threshold);
+        Some(max(l0_flush_stall_threshold, compaction_threshold))
+    }
+
+    fn get_l0_flush_wait_upload(&self) -> bool {
+        let tenant_conf = self.tenant_conf.load();
+        tenant_conf
+            .tenant_conf
+            .l0_flush_wait_upload
+            .unwrap_or(self.conf.default_tenant_conf.l0_flush_wait_upload)
+    }
+
    fn get_image_creation_threshold(&self) -> usize {
        let tenant_conf = self.tenant_conf.load();
        tenant_conf
@@ -2371,6 +2503,7 @@ impl Timeline {
                gate: Gate::default(),

                compaction_lock: tokio::sync::Mutex::default(),
+                compaction_failed: AtomicBool::default(),
                gc_lock: tokio::sync::Mutex::default(),

                standby_horizon: AtomicLsn::new(0),
@@ -2804,12 +2937,10 @@ impl Timeline {
            "initial size calculation",
            // NB: don't log errors here, task_mgr will do that.
            async move {
-                let cancel = task_mgr::shutdown_token();
                self_clone
                    .initial_logical_size_calculation_task(
                        initial_part_end,
                        cancel_wait_for_background_loop_concurrency_limit_semaphore,
-                        cancel,
                        background_ctx,
                    )
                    .await;
@@ -2819,11 +2950,21 @@ impl Timeline {
        );
    }

+    /// # Cancellation
+    ///
+    /// This method is sensitive to `Timeline::cancel`.
+    ///
+    /// It is _not_ sensitive to task_mgr::shutdown_token().
+    ///
+    /// # Cancel-Safety
+    ///
+    /// It does Timeline IO, hence this should be polled to completion because
+    /// we could be leaving in-flight IOs behind, which is safe, but annoying
+    /// to reason about.
    async fn initial_logical_size_calculation_task(
        self: Arc<Self>,
        initial_part_end: Lsn,
        skip_concurrency_limiter: CancellationToken,
-        cancel: CancellationToken,
        background_ctx: RequestContext,
    ) {
        scopeguard::defer! {
@@ -2836,7 +2977,6 @@ impl Timeline {
            let self_ref = &self;
            let skip_concurrency_limiter = &skip_concurrency_limiter;
            async move {
-                let cancel = task_mgr::shutdown_token();
                let wait_for_permit = super::tasks::concurrent_background_tasks_rate_limit_permit(
                    BackgroundLoopKind::InitialLogicalSizeCalculation,
                    background_ctx,
@@ -2850,9 +2990,6 @@ impl Timeline {
                    _ = self_ref.cancel.cancelled() => {
                        return Err(CalculateLogicalSizeError::Cancelled);
                    }
-                    _ = cancel.cancelled() => {
-                        return Err(CalculateLogicalSizeError::Cancelled);
-                    },
                    () = skip_concurrency_limiter.cancelled() => {
                        // Some action that is part of a end user interaction requested logical size
                        // => break out of the rate limit
@@ -2869,6 +3006,14 @@ impl Timeline {
                    crate::metrics::initial_logical_size::START_CALCULATION.retry(circumstances)
                };

+                let io_concurrency = IoConcurrency::spawn_from_conf(
+                    self_ref.conf,
+                    self_ref
+                        .gate
+                        .enter()
+                        .map_err(|_| CalculateLogicalSizeError::Cancelled)?,
+                );
+
                let calculated_size = self_ref
                    .logical_size_calculation_task(
                        initial_part_end,
@@ -2878,7 +3023,11 @@ impl Timeline {
                    .await?;

                self_ref
-                    .trigger_aux_file_size_computation(initial_part_end, background_ctx)
+                    .trigger_aux_file_size_computation(
+                        initial_part_end,
+                        background_ctx,
+                        io_concurrency,
+                    )
                    .await?;

                // TODO: add aux file size to logical size
@@ -2911,22 +3060,18 @@ impl Timeline {
                            )
                            .expect("10min < 1hour"),
                        );
-                        tokio::time::sleep(sleep_duration).await;
+                        tokio::select! {
+                            _ = tokio::time::sleep(sleep_duration) => {}
+                            _ = self.cancel.cancelled() => return ControlFlow::Break(()),
+                        }
                    }
                }
            }
        };

-        let (calculated_size, metrics_guard) = tokio::select! {
-            res = retrying  => {
-                match res {
-                    ControlFlow::Continue(calculated_size) => calculated_size,
-                    ControlFlow::Break(()) => return,
-                }
-            }
-            _ = cancel.cancelled() => {
-                return;
-            }
+        let (calculated_size, metrics_guard) = match retrying.await {
+            ControlFlow::Continue(calculated_size) => calculated_size,
+            ControlFlow::Break(()) => return,
        };

        // we cannot query current_logical_size.current_size() to know the current
@@ -2982,9 +3127,6 @@ impl Timeline {
        receiver
    }

-    /// # Cancel-Safety
-    ///
-    /// This method is cancellation-safe.
    #[instrument(skip_all)]
    async fn logical_size_calculation_task(
        self: &Arc<Self>,
@@ -3002,32 +3144,13 @@ impl Timeline {
            .enter()
            .map_err(|_| CalculateLogicalSizeError::Cancelled)?;

-        let self_calculation = Arc::clone(self);
-
-        let mut calculation = pin!(async {
-            let ctx = ctx.attached_child();
-            self_calculation
-                .calculate_logical_size(lsn, cause, &guard, &ctx)
-                .await
-        });
-
-        tokio::select! {
-            res = &mut calculation => { res }
-            _ = self.cancel.cancelled() => {
-                debug!("cancelling logical size calculation for timeline shutdown");
-                calculation.await
-            }
-        }
+        self.calculate_logical_size(lsn, cause, &guard, ctx).await
    }

    /// Calculate the logical size of the database at the latest LSN.
    ///
    /// NOTE: counted incrementally, includes ancestors. This can be a slow operation,
    /// especially if we need to download remote layers.
-    ///
-    /// # Cancel-Safety
-    ///
-    /// This method is cancellation-safe.
    async fn calculate_logical_size(
        &self,
        up_to_lsn: Lsn,
@@ -3040,7 +3163,10 @@ impl Timeline {
            self.timeline_id, up_to_lsn
        );

-        pausable_failpoint!("timeline-calculate-logical-size-pause");
+        if let Err(()) = pausable_failpoint!("timeline-calculate-logical-size-pause", &self.cancel)
+        {
+            return Err(CalculateLogicalSizeError::Cancelled);
+        }

        // See if we've already done the work for initial size calculation.
        // This is a short-cut for timelines that are mostly unused.
@@ -3473,7 +3599,12 @@ impl Timeline {
            }
        }
        ancestor
-            .wait_lsn(self.ancestor_lsn, WaitLsnWaiter::Timeline(self), ctx)
+            .wait_lsn(
+                self.ancestor_lsn,
+                WaitLsnWaiter::Timeline(self),
+                WaitLsnTimeout::Default,
+                ctx,
+            )
            .await
            .map_err(|e| match e {
                e @ WaitLsnError::Timeout(_) => GetReadyAncestorError::AncestorLsnTimeout(e),
@@ -3593,6 +3724,12 @@ impl Timeline {
        mut layer_flush_start_rx: tokio::sync::watch::Receiver<(u64, Lsn)>,
        ctx: &RequestContext,
    ) {
+        // Subscribe to L0 delta layer updates, for compaction backpressure.
+        let mut watch_l0 = match self.layers.read().await.layer_map() {
+            Ok(lm) => lm.watch_level0_deltas(),
+            Err(Shutdown) => return,
+        };
+
        info!("started flush loop");
        loop {
            tokio::select! {
@@ -3617,43 +3754,68 @@ impl Timeline {
                    return;
                }

-                let timer = self.metrics.flush_time_histo.start_timer();
+                // Break to notify potential waiters as soon as we've flushed the requested LSN. If
+                // more requests have arrived in the meanwhile, we'll resume flushing afterwards.
+                if flushed_to_lsn >= frozen_to_lsn {
+                    break Ok(());
+                }

-                let num_frozen_layers;
-                let frozen_layer_total_size;
-                let layer_to_flush = {
-                    let guard = self.layers.read().await;
-                    let Ok(lm) = guard.layer_map() else {
+                // Fetch the next layer to flush, if any.
+                let (layer, l0_count, frozen_count, frozen_size) = {
+                    let layers = self.layers.read().await;
+                    let Ok(lm) = layers.layer_map() else {
                        info!("dropping out of flush loop for timeline shutdown");
                        return;
                    };
-                    num_frozen_layers = lm.frozen_layers.len();
-                    frozen_layer_total_size = lm
+                    let l0_count = lm.level0_deltas().len();
+                    let frozen_count = lm.frozen_layers.len();
+                    let frozen_size: u64 = lm
                        .frozen_layers
                        .iter()
                        .map(|l| l.estimated_in_mem_size())
-                        .sum::<u64>();
-                    lm.frozen_layers.front().cloned()
-                    // drop 'layers' lock to allow concurrent reads and writes
+                        .sum();
+                    let layer = lm.frozen_layers.front().cloned();
+                    (layer, l0_count, frozen_count, frozen_size)
+                    // drop 'layers' lock
                };
-                let Some(layer_to_flush) = layer_to_flush else {
+                let Some(layer) = layer else {
                    break Ok(());
                };
-                if num_frozen_layers
-                    > std::cmp::max(
-                        self.get_compaction_threshold(),
-                        DEFAULT_COMPACTION_THRESHOLD,
-                    )
-                    && frozen_layer_total_size >= /* 128 MB */ 128000000
-                {
-                    tracing::warn!(
-                        "too many frozen layers: {num_frozen_layers} layers with estimated in-mem size of {frozen_layer_total_size} bytes",
-                    );
-                }
-                match self.flush_frozen_layer(layer_to_flush, ctx).await {
-                    Ok(this_layer_to_lsn) => {
-                        flushed_to_lsn = std::cmp::max(flushed_to_lsn, this_layer_to_lsn);
+
+                // Stall flushes to backpressure if compaction can't keep up. This is propagated up
+                // to WAL ingestion by having ephemeral layer rolls wait for flushes.
+                //
+                // NB: the compaction loop only checks `compaction_threshold` every 20 seconds, so
+                // we can end up stalling before compaction even starts. Consider making it more
+                // responsive (e.g. via `watch_level0_deltas`).
+                if let Some(stall_threshold) = self.get_l0_flush_stall_threshold() {
+                    if l0_count >= stall_threshold {
+                        warn!(
+                            "stalling layer flushes for compaction backpressure at {l0_count} \
+                            L0 layers ({frozen_count} frozen layers with {frozen_size} bytes)"
+                        );
+                        let stall_timer = self
+                            .metrics
+                            .flush_delay_histo
+                            .start_timer()
+                            .record_on_drop();
+                        tokio::select! {
+                            result = watch_l0.wait_for(|l0| *l0 < stall_threshold) => {
+                                if let Ok(l0) = result.as_deref() {
+                                    let delay = stall_timer.elapsed().as_secs_f64();
+                                    info!("resuming layer flushes at {l0} L0 layers after {delay:.3}s");
+                                }
+                            },
+                            _ = self.cancel.cancelled() => {},
+                        }
+                        continue; // check again
                    }
+                }
+
+                // Flush the layer.
+                let flush_timer = self.metrics.flush_time_histo.start_timer();
+                match self.flush_frozen_layer(layer, ctx).await {
+                    Ok(layer_lsn) => flushed_to_lsn = max(flushed_to_lsn, layer_lsn),
                    Err(FlushLayerError::Cancelled) => {
                        info!("dropping out of flush loop for timeline shutdown");
                        return;
@@ -3667,7 +3829,30 @@ impl Timeline {
                        break err.map(|_| ());
                    }
                }
-                timer.stop_and_record();
+                let flush_duration = flush_timer.stop_and_record();
+
+                // Delay the next flush to backpressure if compaction can't keep up. We delay by the
+                // flush duration such that the flush takes 2x as long. This is propagated up to WAL
+                // ingestion by having ephemeral layer rolls wait for flushes.
+                if let Some(delay_threshold) = self.get_l0_flush_delay_threshold() {
+                    if l0_count >= delay_threshold {
+                        let delay = flush_duration.as_secs_f64();
+                        info!(
+                            "delaying layer flush by {delay:.3}s for compaction backpressure at \
+                            {l0_count} L0 layers ({frozen_count} frozen layers with {frozen_size} bytes)"
+                        );
+                        let _delay_timer = self
+                            .metrics
+                            .flush_delay_histo
+                            .start_timer()
+                            .record_on_drop();
+                        tokio::select! {
+                            _ = tokio::time::sleep(flush_duration) => {},
+                            _ = watch_l0.wait_for(|l0| *l0 < delay_threshold) => {},
+                            _ = self.cancel.cancelled() => {},
+                        }
+                    }
+                }
            };

            // Unsharded tenants should never advance their LSN beyond the end of the
@@ -3871,21 +4056,24 @@ impl Timeline {

        // Backpressure mechanism: wait with continuation of the flush loop until we have uploaded all layer files.
        // This makes us refuse ingest until the new layers have been persisted to the remote
-        let start = Instant::now();
-        self.remote_client
-            .wait_completion()
-            .await
-            .map_err(|e| match e {
-                WaitCompletionError::UploadQueueShutDownOrStopped
-                | WaitCompletionError::NotInitialized(
-                    NotInitialized::ShuttingDown | NotInitialized::Stopped,
-                ) => FlushLayerError::Cancelled,
-                WaitCompletionError::NotInitialized(NotInitialized::Uninitialized) => {
-                    FlushLayerError::Other(anyhow!(e).into())
-                }
-            })?;
-        let duration = start.elapsed().as_secs_f64();
-        self.metrics.flush_wait_upload_time_gauge_add(duration);
+        // TODO: remove this, and rely on l0_flush_{delay,stall}_threshold instead.
+        if self.get_l0_flush_wait_upload() {
+            let start = Instant::now();
+            self.remote_client
+                .wait_completion()
+                .await
+                .map_err(|e| match e {
+                    WaitCompletionError::UploadQueueShutDownOrStopped
+                    | WaitCompletionError::NotInitialized(
+                        NotInitialized::ShuttingDown | NotInitialized::Stopped,
+                    ) => FlushLayerError::Cancelled,
+                    WaitCompletionError::NotInitialized(NotInitialized::Uninitialized) => {
+                        FlushLayerError::Other(anyhow!(e).into())
+                    }
+                })?;
+            let duration = start.elapsed().as_secs_f64();
+            self.metrics.flush_wait_upload_time_gauge_add(duration);
+        }

        // FIXME: between create_delta_layer and the scheduling of the upload in `update_metadata_file`,
        // a compaction can delete the file and then it won't be available for uploads any more.
@@ -4128,6 +4316,7 @@ impl Timeline {

    /// Create image layers for Postgres data. Assumes the caller passes a partition that is not too large,
    /// so that at most one image layer will be produced from this function.
+    #[allow(clippy::too_many_arguments)]
    async fn create_image_layer_for_rel_blocks(
        self: &Arc<Self>,
        partition: &KeySpace,
@@ -4136,6 +4325,7 @@ impl Timeline {
        ctx: &RequestContext,
        img_range: Range<Key>,
        start: Key,
+        io_concurrency: IoConcurrency,
    ) -> Result<ImageLayerCreationOutcome, CreateImageLayersError> {
        let mut wrote_keys = false;

@@ -4164,7 +4354,12 @@ impl Timeline {
                    || (last_key_in_range && key_request_accum.raw_size() > 0)
                {
                    let results = self
-                        .get_vectored(key_request_accum.consume_keyspace(), lsn, ctx)
+                        .get_vectored(
+                            key_request_accum.consume_keyspace(),
+                            lsn,
+                            io_concurrency.clone(),
+                            ctx,
+                        )
                        .await?;

                    if self.cancel.is_cancelled() {
@@ -4243,9 +4438,10 @@ impl Timeline {
        img_range: Range<Key>,
        mode: ImageLayerCreationMode,
        start: Key,
+        io_concurrency: IoConcurrency,
    ) -> Result<ImageLayerCreationOutcome, CreateImageLayersError> {
        // Metadata keys image layer creation.
-        let mut reconstruct_state = ValuesReconstructState::default();
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency);
        let begin = Instant::now();
        let data = self
            .get_vectored_impl(partition.clone(), lsn, &mut reconstruct_state, ctx)
@@ -4462,6 +4658,13 @@ impl Timeline {
                )))
            });

+            let io_concurrency = IoConcurrency::spawn_from_conf(
+                self.conf,
+                self.gate
+                    .enter()
+                    .map_err(|_| CreateImageLayersError::Cancelled)?,
+            );
+
            if !compact_metadata {
                let ImageLayerCreationOutcome {
                    image,
@@ -4474,6 +4677,7 @@ impl Timeline {
                        ctx,
                        img_range,
                        start,
+                        io_concurrency,
                    )
                    .await?;

@@ -4492,6 +4696,7 @@ impl Timeline {
                        img_range,
                        mode,
                        start,
+                        io_concurrency,
                    )
                    .await?;
                start = next_start_key;
@@ -5759,13 +5964,14 @@ impl Timeline {
        self: &Arc<Timeline>,
        lsn: Lsn,
        ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
    ) -> anyhow::Result<Vec<(Key, Bytes)>> {
        let mut all_data = Vec::new();
        let guard = self.layers.read().await;
        for layer in guard.layer_map()?.iter_historic_layers() {
            if !layer.is_delta() && layer.image_layer_lsn() == lsn {
                let layer = guard.get_from_desc(&layer);
-                let mut reconstruct_data = ValuesReconstructState::default();
+                let mut reconstruct_data = ValuesReconstructState::new(io_concurrency.clone());
                layer
                    .get_values_reconstruct_data(
                        KeySpace::single(Key::MIN..Key::MAX),
@@ -5774,8 +5980,9 @@ impl Timeline {
                        ctx,
                    )
                    .await?;
-                for (k, v) in reconstruct_data.keys {
-                    all_data.push((k, v?.img.unwrap().1));
+                for (k, v) in std::mem::take(&mut reconstruct_data.keys) {
+                    let v = v.collect_pending_ios().await?;
+                    all_data.push((k, v.img.unwrap().1));
                }
            }
        }
@@ -5896,13 +6103,37 @@ impl TimelineWriter<'_> {
    async fn roll_layer(&mut self, freeze_at: Lsn) -> Result<(), FlushLayerError> {
        let current_size = self.write_guard.as_ref().unwrap().current_size;

+        // If layer flushes are backpressured due to compaction not keeping up, wait for the flush
+        // to propagate the backpressure up into WAL ingestion.
+        let l0_count = self
+            .tl
+            .layers
+            .read()
+            .await
+            .layer_map()?
+            .level0_deltas()
+            .len();
+        let wait_thresholds = [
+            self.get_l0_flush_delay_threshold(),
+            self.get_l0_flush_stall_threshold(),
+        ];
+        let wait_threshold = wait_thresholds.into_iter().flatten().min();
+
        // self.write_guard will be taken by the freezing
-        self.tl
+        let flush_id = self
+            .tl
            .freeze_inmem_layer_at(freeze_at, &mut self.write_guard)
            .await?;

        assert!(self.write_guard.is_none());

+        if let Some(wait_threshold) = wait_threshold {
+            if l0_count >= wait_threshold {
+                info!("layer roll waiting for flush due to compaction backpressure at {l0_count} L0 layers");
+                self.tl.wait_flush_completion(flush_id).await?;
+            }
+        }
+
        if current_size >= self.get_checkpoint_distance() * 2 {
            warn!("Flushed oversized open layer with size {}", current_size)
        }
--- a/pageserver/src/tenant/timeline/compaction.rs
+++ b/pageserver/src/tenant/timeline/compaction.rs
@@ -42,8 +42,8 @@ use crate::tenant::storage_layer::merge_iterator::MergeIterator;
 use crate::tenant::storage_layer::{
    AsLayerDesc, PersistentLayerDesc, PersistentLayerKey, ValueReconstructState,
 };
-use crate::tenant::timeline::ImageLayerCreationOutcome;
 use crate::tenant::timeline::{drop_rlock, DeltaLayerWriter, ImageLayerWriter};
+use crate::tenant::timeline::{ImageLayerCreationOutcome, IoConcurrency};
 use crate::tenant::timeline::{Layer, ResidentLayer};
 use crate::tenant::{gc_block, DeltaLayer, MaybeOffloaded};
 use crate::virtual_file::{MaybeFatalIo, VirtualFile};
@@ -436,12 +436,14 @@ impl KeyHistoryRetention {
        if dry_run {
            return true;
        }
-        let guard = tline.layers.read().await;
-        if !guard.contains_key(key) {
-            return false;
+        let layer_generation;
+        {
+            let guard = tline.layers.read().await;
+            if !guard.contains_key(key) {
+                return false;
+            }
+            layer_generation = guard.get_from_key(key).metadata().generation;
        }
-        let layer_generation = guard.get_from_key(key).metadata().generation;
-        drop(guard);
        if layer_generation == tline.generation {
            info!(
                key=%key,
@@ -2138,6 +2140,11 @@ impl Timeline {
            self.get_gc_compaction_watermark()
        };

+        if compact_below_lsn == Lsn::INVALID {
+            tracing::warn!("no layers to compact with gc: gc_cutoff not generated yet, skipping gc bottom-most compaction");
+            return Ok(vec![]);
+        }
+
        // Split compaction job to about 4GB each
        const GC_COMPACT_MAX_SIZE_MB: u64 = 4 * 1024;
        let sub_compaction_max_job_size_mb =
@@ -2212,6 +2219,12 @@ impl Timeline {
                } else {
                    end
                };
+                let end = if ranges_num == idx + 1 {
+                    // extend the compaction range to the end of the key range if it's the last partition
+                    end.max(job.compact_key_range.end)
+                } else {
+                    end
+                };
                info!(
                    "splitting compaction job: {}..{}, estimated_size={}",
                    start, end, total_size
@@ -2332,6 +2345,11 @@ impl Timeline {
                // each of the retain_lsn. Therefore, if the user-provided `compact_lsn_range.end` is larger than the real gc cutoff, we will use
                // the real cutoff.
                let mut gc_cutoff = if compact_lsn_range.end == Lsn::MAX {
+                    if real_gc_cutoff == Lsn::INVALID {
+                        // If the gc_cutoff is not generated yet, we should not compact anything.
+                        tracing::warn!("no layers to compact with gc: gc_cutoff not generated yet, skipping gc bottom-most compaction");
+                        return Ok(());
+                    }
                    real_gc_cutoff
                } else {
                    compact_lsn_range.end
@@ -2863,7 +2881,7 @@ impl Timeline {
            "produced {} delta layers and {} image layers, {} layers are kept",
            produced_delta_layers_len,
            produced_image_layers_len,
-            layer_selection.len()
+            keep_layers.len()
        );

        // Step 3: Place back to the layer map.
@@ -2909,8 +2927,28 @@ impl Timeline {
        // be batched into `schedule_compaction_update`.
        let disk_consistent_lsn = self.disk_consistent_lsn.load();
        self.schedule_uploads(disk_consistent_lsn, None)?;
+        // If a layer gets rewritten throughout gc-compaction, we need to keep that layer only in `compact_to` instead
+        // of `compact_from`.
+        let compact_from = {
+            let mut compact_from = Vec::new();
+            let mut compact_to_set = HashMap::new();
+            for layer in &compact_to {
+                compact_to_set.insert(layer.layer_desc().key(), layer);
+            }
+            for layer in &layer_selection {
+                if let Some(to) = compact_to_set.get(&layer.layer_desc().key()) {
+                    tracing::info!(
+                        "skipping delete {} because found same layer key at different generation {}",
+                        layer, to
+                    );
+                } else {
+                    compact_from.push(layer.clone());
+                }
+            }
+            compact_from
+        };
        self.remote_client
-            .schedule_compaction_update(&layer_selection, &compact_to)?;
+            .schedule_compaction_update(&compact_from, &compact_to)?;

        drop(gc_lock);

@@ -3170,6 +3208,7 @@ impl TimelineAdaptor {
                ctx,
                key_range.clone(),
                start,
+                IoConcurrency::sequential(),
            )
            .await?;

--- a/pageserver/src/tenant/timeline/import_pgdata.rs
+++ b/pageserver/src/tenant/timeline/import_pgdata.rs
@@ -113,7 +113,7 @@ pub async fn doit(
            match res {
                Ok(_) => break,
                Err(err) => {
-                    info!(?err, "indefintely waiting for pgdata to finish");
+                    info!(?err, "indefinitely waiting for pgdata to finish");
                    if tokio::time::timeout(std::time::Duration::from_secs(10), cancel.cancelled())
                        .await
                        .is_ok()
--- a/pageserver/src/tenant/timeline/import_pgdata/importbucket_client.rs
+++ b/pageserver/src/tenant/timeline/import_pgdata/importbucket_client.rs
@@ -308,7 +308,7 @@ impl ControlFile {
            202107181 => 14,
            202209061 => 15,
            202307071 => 16,
-            /* XXX pg17 */
+            202406281 => 17,
            catversion => {
                anyhow::bail!("unrecognized catalog version {catversion}")
            }
--- a/pageserver/src/tenant/timeline/layer_manager.rs
+++ b/pageserver/src/tenant/timeline/layer_manager.rs
@@ -337,16 +337,45 @@ impl OpenLayerManager {
        compact_to: &[ResidentLayer],
        metrics: &TimelineMetrics,
    ) {
-        // We can simply reuse compact l0 logic. Use a different function name to indicate a different type of layer map modification.
-        self.finish_compact_l0(compact_from, compact_to, metrics)
+        // gc-compaction could contain layer rewrites. We need to delete the old layers and insert the new ones.
+
+        // Match the old layers with the new layers
+        let mut add_layers = HashMap::new();
+        let mut rewrite_layers = HashMap::new();
+        let mut drop_layers = HashMap::new();
+        for layer in compact_from {
+            drop_layers.insert(layer.layer_desc().key(), layer.clone());
+        }
+        for layer in compact_to {
+            if let Some(old_layer) = drop_layers.remove(&layer.layer_desc().key()) {
+                rewrite_layers.insert(layer.layer_desc().key(), (old_layer.clone(), layer.clone()));
+            } else {
+                add_layers.insert(layer.layer_desc().key(), layer.clone());
+            }
+        }
+        let add_layers = add_layers.values().cloned().collect::<Vec<_>>();
+        let drop_layers = drop_layers.values().cloned().collect::<Vec<_>>();
+        let rewrite_layers = rewrite_layers.values().cloned().collect::<Vec<_>>();
+
+        self.rewrite_layers_inner(&rewrite_layers, &drop_layers, &add_layers, metrics);
    }

    /// Called post-compaction when some previous generation image layers were trimmed.
-    pub(crate) fn rewrite_layers(
+    pub fn rewrite_layers(
        &mut self,
        rewrite_layers: &[(Layer, ResidentLayer)],
        drop_layers: &[Layer],
        metrics: &TimelineMetrics,
+    ) {
+        self.rewrite_layers_inner(rewrite_layers, drop_layers, &[], metrics);
+    }
+
+    fn rewrite_layers_inner(
+        &mut self,
+        rewrite_layers: &[(Layer, ResidentLayer)],
+        drop_layers: &[Layer],
+        add_layers: &[ResidentLayer],
+        metrics: &TimelineMetrics,
    ) {
        let mut updates = self.layer_map.batch_update();
        for (old_layer, new_layer) in rewrite_layers {
@@ -382,6 +411,10 @@ impl OpenLayerManager {
        for l in drop_layers {
            Self::delete_historic_layer(l, &mut updates, &mut self.layer_fmgr);
        }
+        for l in add_layers {
+            Self::insert_historic_layer(l.as_ref().clone(), &mut updates, &mut self.layer_fmgr);
+            metrics.record_new_file_metrics(l.layer_desc().file_size);
+        }
        updates.flush();
    }

--- a/pageserver/src/tenant/timeline/walreceiver/connection_manager.rs
+++ b/pageserver/src/tenant/timeline/walreceiver/connection_manager.rs
@@ -164,9 +164,10 @@ pub(super) async fn connection_manager_loop_step(
                    Ok(Some(broker_update)) => connection_manager_state.register_timeline_update(broker_update),
                    Err(status) => {
                        match status.code() {
-                            Code::Unknown if status.message().contains("stream closed because of a broken pipe") || status.message().contains("connection reset") => {
+                            Code::Unknown if status.message().contains("stream closed because of a broken pipe") || status.message().contains("connection reset") || status.message().contains("error reading a body from connection") => {
                                // tonic's error handling doesn't provide a clear code for disconnections: we get
                                // "h2 protocol error: error reading a body from connection: stream closed because of a broken pipe"
+                                // => https://github.com/neondatabase/neon/issues/9562
                                info!("broker disconnected: {status}");
                            },
                            _ => {
@@ -273,7 +274,7 @@ pub(super) async fn connection_manager_loop_step(
                    };

                last_discovery_ts = Some(std::time::Instant::now());
-                debug!("No active connection and no candidates, sending discovery request to the broker");
+                info!("No active connection and no candidates, sending discovery request to the broker");

                // Cancellation safety: we want to send a message to the broker, but publish_one()
                // function can get cancelled by the other select! arm. This is absolutely fine, because
--- a/pageserver/src/tenant/timeline/walreceiver/walreceiver_connection.rs
+++ b/pageserver/src/tenant/timeline/walreceiver/walreceiver_connection.rs
@@ -118,7 +118,7 @@ pub(super) async fn handle_walreceiver_connection(
    cancellation: CancellationToken,
    connect_timeout: Duration,
    ctx: RequestContext,
-    node: NodeId,
+    safekeeper_node: NodeId,
    ingest_batch_size: u64,
 ) -> Result<(), WalReceiverError> {
    debug_assert_current_span_has_tenant_and_timeline_id();
@@ -140,7 +140,7 @@ pub(super) async fn handle_walreceiver_connection(

    let (replication_client, connection) = {
        let mut config = wal_source_connconf.to_tokio_postgres_config();
-        config.application_name(format!("pageserver-{}", node.0).as_str());
+        config.application_name(format!("pageserver-{}", timeline.conf.id.0).as_str());
        config.replication_mode(tokio_postgres::config::ReplicationMode::Physical);
        match time::timeout(connect_timeout, config.connect(postgres::NoTls)).await {
            Ok(client_and_conn) => client_and_conn?,
@@ -162,7 +162,7 @@ pub(super) async fn handle_walreceiver_connection(
        latest_wal_update: Utc::now().naive_utc(),
        streaming_lsn: None,
        commit_lsn: None,
-        node,
+        node: safekeeper_node,
    };
    if let Err(e) = events_sender.send(TaskStateUpdate::Progress(connection_status)) {
        warn!("Wal connection event listener dropped right after connection init, aborting the connection: {e}");
--- a/pageserver/src/tenant/vectored_blob_io.rs
+++ b/pageserver/src/tenant/vectored_blob_io.rs
@@ -35,6 +35,7 @@ use crate::virtual_file::{self, VirtualFile};
 pub struct BlobMeta {
    pub key: Key,
    pub lsn: Lsn,
+    pub will_init: bool,
 }

 /// A view into the vectored blobs read buffer.
@@ -310,7 +311,15 @@ pub enum BlobFlag {
 /// * Iterate over the collected blobs and coalesce them into reads at the end
 pub struct VectoredReadPlanner {
    // Track all the blob offsets. Start offsets must be ordered.
-    blobs: BTreeMap<Key, Vec<(Lsn, u64, u64)>>,
+    // Values in the value tuples are:
+    // (
+    //   lsn of the blob,
+    //   start offset of the blob in the underlying file,
+    //   end offset of the blob in the underlying file,
+    //   whether the blob initializes the page image or not
+    //   see [`pageserver_api::record::NeonWalRecord::will_init`]
+    // )
+    blobs: BTreeMap<Key, Vec<(Lsn, u64, u64, bool)>>,
    // Arguments for previous blob passed into [`VectoredReadPlanner::handle`]
    prev: Option<(Key, Lsn, u64, BlobFlag)>,

@@ -371,12 +380,12 @@ impl VectoredReadPlanner {
        match flag {
            BlobFlag::None => {
                let blobs_for_key = self.blobs.entry(key).or_default();
-                blobs_for_key.push((lsn, start_offset, end_offset));
+                blobs_for_key.push((lsn, start_offset, end_offset, false));
            }
            BlobFlag::ReplaceAll => {
                let blobs_for_key = self.blobs.entry(key).or_default();
                blobs_for_key.clear();
-                blobs_for_key.push((lsn, start_offset, end_offset));
+                blobs_for_key.push((lsn, start_offset, end_offset, true));
            }
            BlobFlag::Ignore => {}
        }
@@ -387,11 +396,17 @@ impl VectoredReadPlanner {
        let mut reads = Vec::new();

        for (key, blobs_for_key) in self.blobs {
-            for (lsn, start_offset, end_offset) in blobs_for_key {
+            for (lsn, start_offset, end_offset, will_init) in blobs_for_key {
                let extended = match &mut current_read_builder {
-                    Some(read_builder) => {
-                        read_builder.extend(start_offset, end_offset, BlobMeta { key, lsn })
-                    }
+                    Some(read_builder) => read_builder.extend(
+                        start_offset,
+                        end_offset,
+                        BlobMeta {
+                            key,
+                            lsn,
+                            will_init,
+                        },
+                    ),
                    None => VectoredReadExtended::No,
                };

@@ -399,7 +414,11 @@ impl VectoredReadPlanner {
                    let next_read_builder = ChunkedVectoredReadBuilder::new(
                        start_offset,
                        end_offset,
-                        BlobMeta { key, lsn },
+                        BlobMeta {
+                            key,
+                            lsn,
+                            will_init,
+                        },
                        self.max_read_size,
                    );

@@ -527,7 +546,7 @@ impl<'a> VectoredBlobReader<'a> {
 pub struct StreamingVectoredReadPlanner {
    read_builder: Option<ChunkedVectoredReadBuilder>,
    // Arguments for previous blob passed into [`StreamingVectoredReadPlanner::handle`]
-    prev: Option<(Key, Lsn, u64)>,
+    prev: Option<(Key, Lsn, u64, bool)>,
    /// Max read size per batch. This is not a strict limit. If there are [0, 100) and [100, 200), while the `max_read_size` is 150,
    /// we will produce a single batch instead of split them.
    max_read_size: u64,
@@ -550,27 +569,47 @@ impl StreamingVectoredReadPlanner {
        }
    }

-    pub fn handle(&mut self, key: Key, lsn: Lsn, offset: u64) -> Option<VectoredRead> {
+    pub fn handle(
+        &mut self,
+        key: Key,
+        lsn: Lsn,
+        offset: u64,
+        will_init: bool,
+    ) -> Option<VectoredRead> {
        // Implementation note: internally lag behind by one blob such that
        // we have a start and end offset when initialising [`VectoredRead`]
-        let (prev_key, prev_lsn, prev_offset) = match self.prev {
+        let (prev_key, prev_lsn, prev_offset, prev_will_init) = match self.prev {
            None => {
-                self.prev = Some((key, lsn, offset));
+                self.prev = Some((key, lsn, offset, will_init));
                return None;
            }
            Some(prev) => prev,
        };

-        let res = self.add_blob(prev_key, prev_lsn, prev_offset, offset, false);
+        let res = self.add_blob(
+            prev_key,
+            prev_lsn,
+            prev_offset,
+            offset,
+            false,
+            prev_will_init,
+        );

-        self.prev = Some((key, lsn, offset));
+        self.prev = Some((key, lsn, offset, will_init));

        res
    }

    pub fn handle_range_end(&mut self, offset: u64) -> Option<VectoredRead> {
-        let res = if let Some((prev_key, prev_lsn, prev_offset)) = self.prev {
-            self.add_blob(prev_key, prev_lsn, prev_offset, offset, true)
+        let res = if let Some((prev_key, prev_lsn, prev_offset, prev_will_init)) = self.prev {
+            self.add_blob(
+                prev_key,
+                prev_lsn,
+                prev_offset,
+                offset,
+                true,
+                prev_will_init,
+            )
        } else {
            None
        };
@@ -587,10 +626,19 @@ impl StreamingVectoredReadPlanner {
        start_offset: u64,
        end_offset: u64,
        is_last_blob_in_read: bool,
+        will_init: bool,
    ) -> Option<VectoredRead> {
        match &mut self.read_builder {
            Some(read_builder) => {
-                let extended = read_builder.extend(start_offset, end_offset, BlobMeta { key, lsn });
+                let extended = read_builder.extend(
+                    start_offset,
+                    end_offset,
+                    BlobMeta {
+                        key,
+                        lsn,
+                        will_init,
+                    },
+                );
                assert_eq!(extended, VectoredReadExtended::Yes);
            }
            None => {
@@ -598,7 +646,11 @@ impl StreamingVectoredReadPlanner {
                    Some(ChunkedVectoredReadBuilder::new_streaming(
                        start_offset,
                        end_offset,
-                        BlobMeta { key, lsn },
+                        BlobMeta {
+                            key,
+                            lsn,
+                            will_init,
+                        },
                    ))
                };
            }
@@ -812,7 +864,7 @@ mod tests {
        let mut planner = StreamingVectoredReadPlanner::new(max_read_size, 1000);
        let mut reads = Vec::new();
        for (key, lsn, offset, _) in blob_descriptions.clone() {
-            reads.extend(planner.handle(key, lsn, offset));
+            reads.extend(planner.handle(key, lsn, offset, false));
        }
        reads.extend(planner.handle_range_end(652 * 1024));

@@ -850,7 +902,7 @@ mod tests {
        let mut planner = StreamingVectoredReadPlanner::new(max_read_size, 2);
        let mut reads = Vec::new();
        for (key, lsn, offset, _) in blob_descriptions.clone() {
-            reads.extend(planner.handle(key, lsn, offset));
+            reads.extend(planner.handle(key, lsn, offset, false));
        }
        reads.extend(planner.handle_range_end(652 * 1024));

@@ -875,7 +927,7 @@ mod tests {
        {
            let mut planner = StreamingVectoredReadPlanner::new(max_read_size, 1);
            let mut reads = Vec::new();
-            reads.extend(planner.handle(key, lsn, 0));
+            reads.extend(planner.handle(key, lsn, 0, false));
            reads.extend(planner.handle_range_end(652 * 1024));
            assert_eq!(reads.len(), 1);
            validate_read(&reads[0], &[(key, lsn, 0, BlobFlag::None)]);
@@ -883,8 +935,8 @@ mod tests {
        {
            let mut planner = StreamingVectoredReadPlanner::new(max_read_size, 1);
            let mut reads = Vec::new();
-            reads.extend(planner.handle(key, lsn, 0));
-            reads.extend(planner.handle(key, lsn, 128 * 1024));
+            reads.extend(planner.handle(key, lsn, 0, false));
+            reads.extend(planner.handle(key, lsn, 128 * 1024, false));
            reads.extend(planner.handle_range_end(652 * 1024));
            assert_eq!(reads.len(), 2);
            validate_read(&reads[0], &[(key, lsn, 0, BlobFlag::None)]);
@@ -893,8 +945,8 @@ mod tests {
        {
            let mut planner = StreamingVectoredReadPlanner::new(max_read_size, 2);
            let mut reads = Vec::new();
-            reads.extend(planner.handle(key, lsn, 0));
-            reads.extend(planner.handle(key, lsn, 128 * 1024));
+            reads.extend(planner.handle(key, lsn, 0, false));
+            reads.extend(planner.handle(key, lsn, 128 * 1024, false));
            reads.extend(planner.handle_range_end(652 * 1024));
            assert_eq!(reads.len(), 1);
            validate_read(
@@ -923,6 +975,7 @@ mod tests {
        let meta = BlobMeta {
            key: Key::MIN,
            lsn: Lsn(0),
+            will_init: false,
        };

        for (idx, (blob, offset)) in blobs.iter().zip(offsets.iter()).enumerate() {
--- a/pageserver/src/utilization.rs
+++ b/pageserver/src/utilization.rs
@@ -49,7 +49,7 @@ pub(crate) fn regenerate(
    };

    // Express a static value for how many shards we may schedule on one node
-    const MAX_SHARDS: u32 = 20000;
+    const MAX_SHARDS: u32 = 5000;

    let mut doc = PageserverUtilization {
        disk_usage_bytes: used,
--- a/pageserver/src/walingest.rs
+++ b/pageserver/src/walingest.rs
@@ -499,7 +499,13 @@ impl WalIngest {

                let content = modification
                    .tline
-                    .get_rel_page_at_lsn(src_rel, blknum, Version::Modified(modification), ctx)
+                    .get_rel_page_at_lsn(
+                        src_rel,
+                        blknum,
+                        Version::Modified(modification),
+                        ctx,
+                        crate::tenant::storage_layer::IoConcurrency::sequential(),
+                    )
                    .await?;
                modification.put_rel_page_image(dst_rel, blknum, content)?;
                num_blocks_copied += 1;
@@ -1489,6 +1495,7 @@ mod tests {
    use super::*;
    use crate::tenant::harness::*;
    use crate::tenant::remote_timeline_client::{remote_initdb_archive_path, INITDB_PATH};
+    use crate::tenant::storage_layer::IoConcurrency;
    use postgres_ffi::RELSEG_SIZE;

    use crate::DEFAULT_PG_VERSION;
@@ -1532,6 +1539,7 @@ mod tests {
    #[tokio::test]
    async fn test_relsize() -> Result<()> {
        let (tenant, ctx) = TenantHarness::create("test_relsize").await?.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_test_timeline(TIMELINE_ID, Lsn(8), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -1599,7 +1607,13 @@ mod tests {
        // Check page contents at each LSN
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x20)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x20)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 0 at 2")
@@ -1607,7 +1621,13 @@ mod tests {

        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x30)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x30)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 0 at 3")
@@ -1615,14 +1635,26 @@ mod tests {

        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x40)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x40)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 0 at 3")
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 1, Version::Lsn(Lsn(0x40)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    1,
+                    Version::Lsn(Lsn(0x40)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 1 at 4")
@@ -1630,21 +1662,39 @@ mod tests {

        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x50)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x50)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 0 at 3")
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 1, Version::Lsn(Lsn(0x50)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    1,
+                    Version::Lsn(Lsn(0x50)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 1 at 4")
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 2, Version::Lsn(Lsn(0x50)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    2,
+                    Version::Lsn(Lsn(0x50)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 2 at 5")
@@ -1667,14 +1717,26 @@ mod tests {
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x60)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x60)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 0 at 3")
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 1, Version::Lsn(Lsn(0x60)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    1,
+                    Version::Lsn(Lsn(0x60)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 1 at 4")
@@ -1689,7 +1751,13 @@ mod tests {
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 2, Version::Lsn(Lsn(0x50)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    2,
+                    Version::Lsn(Lsn(0x50)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 2 at 5")
@@ -1722,14 +1790,26 @@ mod tests {
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 0, Version::Lsn(Lsn(0x70)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    0,
+                    Version::Lsn(Lsn(0x70)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            ZERO_PAGE
        );
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 1, Version::Lsn(Lsn(0x70)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    1,
+                    Version::Lsn(Lsn(0x70)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 1")
@@ -1750,7 +1830,13 @@ mod tests {
        for blk in 2..1500 {
            assert_eq!(
                tline
-                    .get_rel_page_at_lsn(TESTREL_A, blk, Version::Lsn(Lsn(0x80)), &ctx)
+                    .get_rel_page_at_lsn(
+                        TESTREL_A,
+                        blk,
+                        Version::Lsn(Lsn(0x80)),
+                        &ctx,
+                        io_concurrency.clone()
+                    )
                    .instrument(test_span.clone())
                    .await?,
                ZERO_PAGE
@@ -1758,7 +1844,13 @@ mod tests {
        }
        assert_eq!(
            tline
-                .get_rel_page_at_lsn(TESTREL_A, 1500, Version::Lsn(Lsn(0x80)), &ctx)
+                .get_rel_page_at_lsn(
+                    TESTREL_A,
+                    1500,
+                    Version::Lsn(Lsn(0x80)),
+                    &ctx,
+                    io_concurrency.clone()
+                )
                .instrument(test_span.clone())
                .await?,
            test_img("foo blk 1500")
@@ -1851,6 +1943,7 @@ mod tests {
            .await?
            .load()
            .await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
        let tline = tenant
            .create_test_timeline(TIMELINE_ID, Lsn(8), DEFAULT_PG_VERSION, &ctx)
            .await?;
@@ -1903,7 +1996,13 @@ mod tests {
            let data = format!("foo blk {} at {}", blkno, lsn);
            assert_eq!(
                tline
-                    .get_rel_page_at_lsn(TESTREL_A, blkno, Version::Lsn(lsn), &ctx)
+                    .get_rel_page_at_lsn(
+                        TESTREL_A,
+                        blkno,
+                        Version::Lsn(lsn),
+                        &ctx,
+                        io_concurrency.clone()
+                    )
                    .instrument(test_span.clone())
                    .await?,
                test_img(&data)
@@ -1931,7 +2030,13 @@ mod tests {
            let data = format!("foo blk {} at {}", blkno, lsn);
            assert_eq!(
                tline
-                    .get_rel_page_at_lsn(TESTREL_A, blkno, Version::Lsn(Lsn(0x60)), &ctx)
+                    .get_rel_page_at_lsn(
+                        TESTREL_A,
+                        blkno,
+                        Version::Lsn(Lsn(0x60)),
+                        &ctx,
+                        io_concurrency.clone()
+                    )
                    .instrument(test_span.clone())
                    .await?,
                test_img(&data)
@@ -1950,7 +2055,13 @@ mod tests {
            let data = format!("foo blk {} at {}", blkno, lsn);
            assert_eq!(
                tline
-                    .get_rel_page_at_lsn(TESTREL_A, blkno, Version::Lsn(Lsn(0x50)), &ctx)
+                    .get_rel_page_at_lsn(
+                        TESTREL_A,
+                        blkno,
+                        Version::Lsn(Lsn(0x50)),
+                        &ctx,
+                        io_concurrency.clone()
+                    )
                    .instrument(test_span.clone())
                    .await?,
                test_img(&data)
@@ -1987,7 +2098,13 @@ mod tests {
            let data = format!("foo blk {} at {}", blkno, lsn);
            assert_eq!(
                tline
-                    .get_rel_page_at_lsn(TESTREL_A, blkno, Version::Lsn(Lsn(0x80)), &ctx)
+                    .get_rel_page_at_lsn(
+                        TESTREL_A,
+                        blkno,
+                        Version::Lsn(Lsn(0x80)),
+                        &ctx,
+                        io_concurrency.clone()
+                    )
                    .instrument(test_span.clone())
                    .await?,
                test_img(&data)
--- a/pgxn/neon/neon.c
+++ b/pgxn/neon/neon.c
@@ -19,6 +19,7 @@
 #include "access/xlogrecovery.h"
 #endif
 #include "replication/logical.h"
+#include "replication/logicallauncher.h"
 #include "replication/slot.h"
 #include "replication/walsender.h"
 #include "storage/proc.h"
@@ -434,6 +435,15 @@ _PG_init(void)

 	restore_running_xacts_callback = RestoreRunningXactsFromClog;

+	DefineCustomBoolVariable(
+							"neon.disable_logical_replication_subscribers",
+							"Disables incomming logical replication",
+							NULL,
+							&disable_logical_replication_subscribers,
+							false,
+							PGC_SIGHUP,
+							0,
+							NULL, NULL, NULL);

 	DefineCustomBoolVariable(
 							"neon.allow_replica_misconfig",
--- a/poetry.lock
+++ b/poetry.lock
@@ -1,4 +1,4 @@
-# This file is automatically @generated by Poetry 1.8.5 and should not be changed by hand.
+# This file is automatically @generated by Poetry 2.0.1 and should not be changed by hand.

 [[package]]
 name = "aiohappyeyeballs"
@@ -6,6 +6,7 @@ version = "2.3.5"
 description = "Happy Eyeballs for asyncio"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "aiohappyeyeballs-2.3.5-py3-none-any.whl", hash = "sha256:4d6dea59215537dbc746e93e779caea8178c866856a721c9c660d7a5a7b8be03"},
    {file = "aiohappyeyeballs-2.3.5.tar.gz", hash = "sha256:6fa48b9f1317254f122a07a131a86b71ca6946ca989ce6326fff54a99a920105"},
@@ -17,6 +18,7 @@ version = "3.10.11"
 description = "Async http client/server framework (asyncio)"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "aiohttp-3.10.11-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:5077b1a5f40ffa3ba1f40d537d3bec4383988ee51fbba6b74aa8fb1bc466599e"},
    {file = "aiohttp-3.10.11-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:8d6a14a4d93b5b3c2891fca94fa9d41b2322a68194422bef0dd5ec1e57d7d298"},
@@ -128,6 +130,7 @@ version = "1.4.0"
 description = "Postgres integration with asyncio."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "aiopg-1.4.0-py3-none-any.whl", hash = "sha256:aea46e8aff30b039cfa818e6db4752c97656e893fc75e5a5dc57355a9e9dedbd"},
    {file = "aiopg-1.4.0.tar.gz", hash = "sha256:116253bef86b4d954116716d181e9a0294037f266718b2e1c9766af995639d71"},
@@ -146,6 +149,7 @@ version = "1.3.1"
 description = "aiosignal: a list of registered asynchronous callbacks"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "aiosignal-1.3.1-py3-none-any.whl", hash = "sha256:f8376fb07dd1e86a584e4fcdec80b36b7f81aac666ebc724e2c090300dd83b17"},
    {file = "aiosignal-1.3.1.tar.gz", hash = "sha256:54cd96e15e1649b75d6c87526a6ff0b6c1b0dd3459f43d9ca11d48c339b68cfc"},
@@ -160,6 +164,7 @@ version = "2.13.2"
 description = "Allure pytest integration"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "allure-pytest-2.13.2.tar.gz", hash = "sha256:22243159e8ec81ce2b5254b4013802198821b1b42f118f69d4a289396607c7b3"},
    {file = "allure_pytest-2.13.2-py3-none-any.whl", hash = "sha256:17de9dbee7f61c8e66a5b5e818b00e419dbcea44cb55c24319401ba813220690"},
@@ -175,6 +180,7 @@ version = "2.13.2"
 description = "Common module for integrate allure with python-based frameworks"
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "allure-python-commons-2.13.2.tar.gz", hash = "sha256:8a03681330231b1deadd86b97ff68841c6591320114ae638570f1ed60d7a2033"},
    {file = "allure_python_commons-2.13.2-py3-none-any.whl", hash = "sha256:2bb3646ec3fbf5b36d178a5e735002bc130ae9f9ba80f080af97d368ba375051"},
@@ -190,6 +196,7 @@ version = "0.6.0"
 description = "Reusable constraint types to use with typing.Annotated"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "annotated_types-0.6.0-py3-none-any.whl", hash = "sha256:0641064de18ba7a25dee8f96403ebc39113d0cb953a01429249d5c7564666a43"},
    {file = "annotated_types-0.6.0.tar.gz", hash = "sha256:563339e807e53ffd9c267e99fc6d9ea23eb8443c08f112651963e24e22f84a5d"},
@@ -201,6 +208,7 @@ version = "4.13.1"
 description = "ANTLR 4.13.1 runtime for Python 3"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "antlr4-python3-runtime-4.13.1.tar.gz", hash = "sha256:3cd282f5ea7cfb841537fe01f143350fdb1c0b1ce7981443a2fa8513fddb6d1a"},
    {file = "antlr4_python3_runtime-4.13.1-py3-none-any.whl", hash = "sha256:78ec57aad12c97ac039ca27403ad61cb98aaec8a3f9bb8144f889aa0fa28b943"},
@@ -212,6 +220,7 @@ version = "4.3.0"
 description = "High level compatibility layer for multiple asynchronous event loop implementations"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "anyio-4.3.0-py3-none-any.whl", hash = "sha256:048e05d0f6caeed70d731f3db756d35dcc1f35747c8c403364a8332c630441b8"},
    {file = "anyio-4.3.0.tar.gz", hash = "sha256:f75253795a87df48568485fd18cdd2a3fa5c4f7c5be8e5e36637733fce06fed6"},
@@ -232,6 +241,7 @@ version = "4.0.3"
 description = "Timeout context manager for asyncio programs"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "async-timeout-4.0.3.tar.gz", hash = "sha256:4640d96be84d82d02ed59ea2b7105a0f7b33abe8703703cd0ab0bf87c427522f"},
    {file = "async_timeout-4.0.3-py3-none-any.whl", hash = "sha256:7405140ff1230c310e51dc27b3145b9092d659ce68ff733fb0cefe3ee42be028"},
@@ -243,6 +253,7 @@ version = "0.30.0"
 description = "An asyncio PostgreSQL driver"
 optional = false
 python-versions = ">=3.8.0"
+groups = ["main"]
 files = [
    {file = "asyncpg-0.30.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bfb4dd5ae0699bad2b233672c8fc5ccbd9ad24b89afded02341786887e37927e"},
    {file = "asyncpg-0.30.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:dc1f62c792752a49f88b7e6f774c26077091b44caceb1983509edc18a2222ec0"},
@@ -306,6 +317,7 @@ version = "21.4.0"
 description = "Classes Without Boilerplate"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+groups = ["main"]
 files = [
    {file = "attrs-21.4.0-py2.py3-none-any.whl", hash = "sha256:2d27e3784d7a565d36ab851fe94887c5eccd6a463168875832a1be79c82828b4"},
    {file = "attrs-21.4.0.tar.gz", hash = "sha256:626ba8234211db98e869df76230a137c4c40a12d72445c45d5f5b716f076e2fd"},
@@ -323,6 +335,7 @@ version = "1.88.0"
 description = "AWS SAM Translator is a library that transform SAM templates into AWS CloudFormation templates"
 optional = false
 python-versions = "!=4.0,<=4.0,>=3.8"
+groups = ["main"]
 files = [
    {file = "aws_sam_translator-1.88.0-py3-none-any.whl", hash = "sha256:aa93d498d8de3fb3d485c316155b1628144b823bbc176099a20de06df666fcac"},
    {file = "aws_sam_translator-1.88.0.tar.gz", hash = "sha256:e77c65f3488566122277accd44a0f1ec018e37403e0d5fe25120d96e537e91a7"},
@@ -343,6 +356,7 @@ version = "2.10.0"
 description = "The AWS X-Ray SDK for Python (the SDK) enables Python developers to record and emit information from within their applications to the AWS X-Ray service."
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "aws-xray-sdk-2.10.0.tar.gz", hash = "sha256:9b14924fd0628cf92936055864655354003f0b1acc3e1c3ffde6403d0799dd7a"},
    {file = "aws_xray_sdk-2.10.0-py2.py3-none-any.whl", hash = "sha256:7551e81a796e1a5471ebe84844c40e8edf7c218db33506d046fec61f7495eda4"},
@@ -358,6 +372,7 @@ version = "2.2.1"
 description = "Function decoration for backoff and retry"
 optional = false
 python-versions = ">=3.7,<4.0"
+groups = ["main"]
 files = [
    {file = "backoff-2.2.1-py3-none-any.whl", hash = "sha256:63579f9a0628e06278f7e47b7d7d5b6ce20dc65c5e96a6f3ca99a6adca0396e8"},
    {file = "backoff-2.2.1.tar.gz", hash = "sha256:03f829f5bb1923180821643f8753b0502c3b682293992485b0eef2807afa5cba"},
@@ -369,6 +384,7 @@ version = "1.34.11"
 description = "The AWS SDK for Python"
 optional = false
 python-versions = ">= 3.8"
+groups = ["main"]
 files = [
    {file = "boto3-1.34.11-py3-none-any.whl", hash = "sha256:1af021e0c6e3040e8de66d403e963566476235bb70f9a8e3f6784813ac2d8026"},
    {file = "boto3-1.34.11.tar.gz", hash = "sha256:31c130a40ec0631059b77d7e87f67ad03ff1685a5b37638ac0c4687026a3259d"},
@@ -388,6 +404,7 @@ version = "1.26.16"
 description = "Type annotations for boto3 1.26.16 generated with mypy-boto3-builder 7.11.11"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "boto3-stubs-1.26.16.tar.gz", hash = "sha256:618253ae19f1480785759bcaee8c8b10ed3fc037027247c26a3461a50f58406d"},
    {file = "boto3_stubs-1.26.16-py3-none-any.whl", hash = "sha256:8cf2925bc3e1349c93eb0f49c1061affc5ca314d69eeb335349037969d0787ed"},
@@ -732,6 +749,7 @@ version = "1.34.11"
 description = "Low-level, data-driven core of boto 3."
 optional = false
 python-versions = ">= 3.8"
+groups = ["main"]
 files = [
    {file = "botocore-1.34.11-py3-none-any.whl", hash = "sha256:1ff1398b6ea670e1c01ac67a33af3da854f8e700d3528289c04f319c330d8250"},
    {file = "botocore-1.34.11.tar.gz", hash = "sha256:51905c3d623c60df5dc5794387de7caf886d350180a01a3dfa762e903edb45a9"},
@@ -751,6 +769,7 @@ version = "1.27.38"
 description = "Type annotations for botocore 1.27.38 generated with mypy-boto3-builder 7.10.1"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "botocore-stubs-1.27.38.tar.gz", hash = "sha256:408e8b86b5d171b58f81c74ca9d3b5317a5a8e2d3bc2073aa841ac13b8939e56"},
    {file = "botocore_stubs-1.27.38-py3-none-any.whl", hash = "sha256:7add7641e9a479a9c8366893bb522fd9ca3d58714201e43662a200a148a1bc38"},
@@ -765,6 +784,7 @@ version = "2024.7.4"
 description = "Python package for providing Mozilla's CA Bundle."
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "certifi-2024.7.4-py3-none-any.whl", hash = "sha256:c198e21b1289c2ab85ee4e67bb4b4ef3ead0892059901a8d5b622f24a1101e90"},
    {file = "certifi-2024.7.4.tar.gz", hash = "sha256:5a1e7645bc0ec61a09e26c36f6106dd4cf40c6db3a1fb6352b0244e7fb057c7b"},
@@ -776,6 +796,7 @@ version = "1.17.1"
 description = "Foreign Function Interface for Python calling C code."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "cffi-1.17.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:df8b1c11f177bc2313ec4b2d46baec87a5f3e71fc8b45dab2ee7cae86d9aba14"},
    {file = "cffi-1.17.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:8f2cdc858323644ab277e9bb925ad72ae0e67f69e804f4898c070998d50b1a67"},
@@ -855,6 +876,7 @@ version = "0.87.1"
 description = "Checks CloudFormation templates for practices and behaviour that could potentially be improved"
 optional = false
 python-versions = "!=4.0,<=4.0,>=3.8"
+groups = ["main"]
 files = [
    {file = "cfn_lint-0.87.1-py3-none-any.whl", hash = "sha256:d450f450635fc223b6f66880ccac52a5fd1a52966fa1705f1ba52b88dfed3071"},
    {file = "cfn_lint-0.87.1.tar.gz", hash = "sha256:b3ce9d3e5e0eadcea5d584c8ccaa00bf2a990a36a64d7ffd8683bc60b7e4f06f"},
@@ -878,6 +900,7 @@ version = "2.1.0"
 description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet."
 optional = false
 python-versions = ">=3.6.0"
+groups = ["main"]
 files = [
    {file = "charset-normalizer-2.1.0.tar.gz", hash = "sha256:575e708016ff3a5e3681541cb9d79312c416835686d054a23accb873b254f413"},
    {file = "charset_normalizer-2.1.0-py3-none-any.whl", hash = "sha256:5189b6f22b01957427f35b6a08d9a0bc45b46d3788ef5a92e978433c7a35f8a5"},
@@ -892,6 +915,7 @@ version = "8.1.3"
 description = "Composable command line interface toolkit"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "click-8.1.3-py3-none-any.whl", hash = "sha256:bb4d8133cb15a609f44e8213d9b391b0809795062913b383c62be0ee95b1db48"},
    {file = "click-8.1.3.tar.gz", hash = "sha256:7682dc8afb30297001674575ea00d1814d808d6a36af415a82bd481d37ba7b8e"},
@@ -906,6 +930,7 @@ version = "0.7.17"
 description = "ClickHouse Database Core Driver for Python, Pandas, and Superset"
 optional = false
 python-versions = "~=3.8"
+groups = ["main"]
 files = [
    {file = "clickhouse-connect-0.7.17.tar.gz", hash = "sha256:854f1f9f3e024e7f89ae5d57cd3289d7a4c3dc91a9f24c4d233014f0ea19cb2d"},
    {file = "clickhouse_connect-0.7.17-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:aca36f5f28be1ada2981fce87724bbf451f267c918015baec59e527de3c9c882"},
@@ -996,6 +1021,8 @@ version = "0.4.5"
 description = "Cross-platform colored terminal text."
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+groups = ["main"]
+markers = "sys_platform == \"win32\" or platform_system == \"Windows\""
 files = [
    {file = "colorama-0.4.5-py2.py3-none-any.whl", hash = "sha256:854bf444933e37f5824ae7bfc1e98d5bce2ebe4160d46b5edf346a89358e99da"},
    {file = "colorama-0.4.5.tar.gz", hash = "sha256:e6c6b4334fc50988a639d9b98aa429a0b57da6e17b9a44f0451f930b6967b7a4"},
@@ -1007,6 +1034,7 @@ version = "43.0.1"
 description = "cryptography is a package which provides cryptographic recipes and primitives to Python developers."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "cryptography-43.0.1-cp37-abi3-macosx_10_9_universal2.whl", hash = "sha256:8385d98f6a3bf8bb2d65a73e17ed87a3ba84f6991c155691c51112075f9ffc5d"},
    {file = "cryptography-43.0.1-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:27e613d7077ac613e399270253259d9d53872aaf657471473ebfc9a52935c062"},
@@ -1056,6 +1084,7 @@ version = "7.1.0"
 description = "A Python library for the Docker Engine API."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "docker-7.1.0-py3-none-any.whl", hash = "sha256:c96b93b7f0a746f9e77d325bcfb87422a3d8bd4f03136ae8a85b37f1898d5fc0"},
    {file = "docker-7.1.0.tar.gz", hash = "sha256:ad8c70e6e3f8926cb8a92619b832b4ea5299e2831c14284663184e200546fa6c"},
@@ -1078,6 +1107,7 @@ version = "1.9.0"
 description = "execnet: rapid multi-Python deployment"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+groups = ["main"]
 files = [
    {file = "execnet-1.9.0-py2.py3-none-any.whl", hash = "sha256:a295f7cc774947aac58dde7fdc85f4aa00c42adf5d8f5468fc630c1acf30a142"},
    {file = "execnet-1.9.0.tar.gz", hash = "sha256:8f694f3ba9cc92cab508b152dcfe322153975c29bda272e2fd7f3f00f36e47c5"},
@@ -1092,6 +1122,7 @@ version = "2.2.5"
 description = "A simple framework for building complex web applications."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "Flask-2.2.5-py3-none-any.whl", hash = "sha256:58107ed83443e86067e41eff4631b058178191a355886f8e479e347fa1285fdf"},
    {file = "Flask-2.2.5.tar.gz", hash = "sha256:edee9b0a7ff26621bd5a8c10ff484ae28737a2410d99b0bb9a6850c7fb977aa0"},
@@ -1113,6 +1144,7 @@ version = "5.0.0"
 description = "A Flask extension adding a decorator for CORS support"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "Flask_Cors-5.0.0-py2.py3-none-any.whl", hash = "sha256:b9e307d082a9261c100d8fb0ba909eec6a228ed1b60a8315fd85f783d61910bc"},
    {file = "flask_cors-5.0.0.tar.gz", hash = "sha256:5aadb4b950c4e93745034594d9f3ea6591f734bb3662e16e255ffbf5e89c88ef"},
@@ -1127,6 +1159,7 @@ version = "1.5.0"
 description = "A list-like structure which implements collections.abc.MutableSequence"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "frozenlist-1.5.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:5b6a66c18b5b9dd261ca98dffcb826a525334b2f29e7caa54e182255c5f6a65a"},
    {file = "frozenlist-1.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d1b3eb7b05ea246510b43a7e53ed1653e55c2121019a97e60cad7efb881a97bb"},
@@ -1228,6 +1261,7 @@ version = "3.2.1"
 description = "GraphQL implementation for Python, a port of GraphQL.js, the JavaScript reference implementation for GraphQL."
 optional = false
 python-versions = ">=3.6,<4"
+groups = ["main"]
 files = [
    {file = "graphql-core-3.2.1.tar.gz", hash = "sha256:9d1bf141427b7d54be944587c8349df791ce60ade2e3cccaf9c56368c133c201"},
    {file = "graphql_core-3.2.1-py3-none-any.whl", hash = "sha256:f83c658e4968998eed1923a2e3e3eddd347e005ac0315fbb7ca4d70ea9156323"},
@@ -1239,6 +1273,7 @@ version = "0.14.0"
 description = "A pure-Python, bring-your-own-I/O implementation of HTTP/1.1"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "h11-0.14.0-py3-none-any.whl", hash = "sha256:e3fe4ac4b851c468cc8363d500db52c2ead036020723024a109d37346efaa761"},
    {file = "h11-0.14.0.tar.gz", hash = "sha256:8f19fbbe99e72420ff35c00b27a34cb9937e902a8b810e2c88300c6f0a3b699d"},
@@ -1247,27 +1282,33 @@ files = [
 [[package]]
 name = "h2"
 version = "4.1.0"
-description = "HTTP/2 State-Machine based protocol implementation"
+description = "Pure-Python HTTP/2 protocol implementation"
 optional = false
-python-versions = ">=3.6.1"
-files = [
-    {file = "h2-4.1.0-py3-none-any.whl", hash = "sha256:03a46bcf682256c95b5fd9e9a99c1323584c3eec6440d379b9903d709476bc6d"},
-    {file = "h2-4.1.0.tar.gz", hash = "sha256:a83aca08fbe7aacb79fec788c9c0bac936343560ed9ec18b82a13a12c28d2abb"},
-]
+python-versions = ">=3.9"
+groups = ["main"]
+files = []
+develop = false

 [package.dependencies]
-hpack = ">=4.0,<5"
-hyperframe = ">=6.0,<7"
+hpack = ">=4.1,<5"
+hyperframe = ">=6.1,<7"
+
+[package.source]
+type = "git"
+url = "https://github.com/python-hyper/h2"
+reference = "HEAD"
+resolved_reference = "0b98b244b5fd1fe96100ac14905417a3b70a4286"

 [[package]]
 name = "hpack"
-version = "4.0.0"
-description = "Pure-Python HPACK header compression"
+version = "4.1.0"
+description = "Pure-Python HPACK header encoding"
 optional = false
-python-versions = ">=3.6.1"
+python-versions = ">=3.9"
+groups = ["main"]
 files = [
-    {file = "hpack-4.0.0-py3-none-any.whl", hash = "sha256:84a076fad3dc9a9f8063ccb8041ef100867b1878b25ef0ee63847a5d53818a6c"},
-    {file = "hpack-4.0.0.tar.gz", hash = "sha256:fc41de0c63e687ebffde81187a948221294896f6bdc0ae2312708df339430095"},
+    {file = "hpack-4.1.0-py3-none-any.whl", hash = "sha256:157ac792668d995c657d93111f46b4535ed114f0c9c8d672271bbec7eae1b496"},
+    {file = "hpack-4.1.0.tar.gz", hash = "sha256:ec5eca154f7056aa06f196a557655c5b009b382873ac8d1e66e79e87535f1dca"},
 ]

 [[package]]
@@ -1276,6 +1317,7 @@ version = "1.0.3"
 description = "A minimal low-level HTTP client."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "httpcore-1.0.3-py3-none-any.whl", hash = "sha256:9a6a501c3099307d9fd76ac244e08503427679b1e81ceb1d922485e2f2462ad2"},
    {file = "httpcore-1.0.3.tar.gz", hash = "sha256:5c0f9546ad17dac4d0772b0808856eb616eb8b48ce94f49ed819fd6982a8a544"},
@@ -1297,6 +1339,7 @@ version = "0.26.0"
 description = "The next generation HTTP client."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "httpx-0.26.0-py3-none-any.whl", hash = "sha256:8915f5a3627c4d47b73e8202457cb28f1266982d1159bd5779d86a80c0eab1cd"},
    {file = "httpx-0.26.0.tar.gz", hash = "sha256:451b55c30d5185ea6b23c2c793abf9bb237d2a7dfb901ced6ff69ad37ec1dfaf"},
@@ -1318,13 +1361,14 @@ socks = ["socksio (==1.*)"]

 [[package]]
 name = "hyperframe"
-version = "6.0.1"
-description = "HTTP/2 framing layer for Python"
+version = "6.1.0"
+description = "Pure-Python HTTP/2 framing"
 optional = false
-python-versions = ">=3.6.1"
+python-versions = ">=3.9"
+groups = ["main"]
 files = [
-    {file = "hyperframe-6.0.1-py3-none-any.whl", hash = "sha256:0ec6bafd80d8ad2195c4f03aacba3a8265e57bc4cff261e802bf39970ed02a15"},
-    {file = "hyperframe-6.0.1.tar.gz", hash = "sha256:ae510046231dc8e9ecb1a6586f63d2347bf4c8905914aa84ba585ae85f28a914"},
+    {file = "hyperframe-6.1.0-py3-none-any.whl", hash = "sha256:b03380493a519fce58ea5af42e4a42317bf9bd425596f7a0835ffce80f1a42e5"},
+    {file = "hyperframe-6.1.0.tar.gz", hash = "sha256:f630908a00854a7adeabd6382b43923a4c4cd4b821fcb527e6ab9e15382a3b08"},
 ]

 [[package]]
@@ -1333,6 +1377,7 @@ version = "3.7"
 description = "Internationalized Domain Names in Applications (IDNA)"
 optional = false
 python-versions = ">=3.5"
+groups = ["main"]
 files = [
    {file = "idna-3.7-py3-none-any.whl", hash = "sha256:82fee1fc78add43492d3a1898bfa6d8a904cc97d8427f683ed8e798d07761aa0"},
    {file = "idna-3.7.tar.gz", hash = "sha256:028ff3aadf0609c1fd278d8ea3089299412a7a8b9bd005dd08b9f8285bcb5cfc"},
@@ -1344,6 +1389,7 @@ version = "1.1.1"
 description = "iniconfig: brain-dead simple config-ini parsing"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "iniconfig-1.1.1-py2.py3-none-any.whl", hash = "sha256:011e24c64b7f47f6ebd835bb12a743f2fbe9a26d4cecaa7f53bc4f35ee9da8b3"},
    {file = "iniconfig-1.1.1.tar.gz", hash = "sha256:bc3af051d7d14b2ee5ef9969666def0cd1a000e121eaea580d4a313df4b37f32"},
@@ -1355,6 +1401,7 @@ version = "2.1.2"
 description = "Safely pass data to untrusted environments and back."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "itsdangerous-2.1.2-py3-none-any.whl", hash = "sha256:2c2349112351b88699d8d4b6b075022c0808887cb7ad10069318a8b0bc88db44"},
    {file = "itsdangerous-2.1.2.tar.gz", hash = "sha256:5dbbc68b317e5e42f327f9021763545dc3fc3bfe22e6deb96aaf1fc38874156a"},
@@ -1366,6 +1413,7 @@ version = "3.1.5"
 description = "A very fast and expressive template engine."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "jinja2-3.1.5-py3-none-any.whl", hash = "sha256:aba0f4dc9ed8013c424088f68a5c226f7d6097ed89b246d7749c2ec4175c6adb"},
    {file = "jinja2-3.1.5.tar.gz", hash = "sha256:8fefff8dc3034e27bb80d67c671eb8a9bc424c0ef4c0826edbff304cceff43bb"},
@@ -1383,6 +1431,7 @@ version = "1.0.1"
 description = "JSON Matching Expressions"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "jmespath-1.0.1-py3-none-any.whl", hash = "sha256:02e2e4cc71b5bcab88332eebf907519190dd9e6e82107fa7f83b1003a6252980"},
    {file = "jmespath-1.0.1.tar.gz", hash = "sha256:90261b206d6defd58fdd5e85f478bf633a2901798906be2ad389150c5c60edbe"},
@@ -1394,6 +1443,7 @@ version = "0.9.0"
 description = "The ultimate Python library for JOSE RFCs, including JWS, JWE, JWK, JWA, JWT"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "joserfc-0.9.0-py3-none-any.whl", hash = "sha256:4026bdbe2c196cd40574e916fa1e28874d99649412edaab0e373dec3077153fb"},
    {file = "joserfc-0.9.0.tar.gz", hash = "sha256:eebca7f587b1761ce43a98ffd5327f2b600b9aa5bb0a77b947687f503ad43bc0"},
@@ -1411,6 +1461,7 @@ version = "1.2.3"
 description = "Generate source code for Python classes from a JSON schema."
 optional = false
 python-versions = ">= 2.7"
+groups = ["main"]
 files = [
    {file = "jschema_to_python-1.2.3-py3-none-any.whl", hash = "sha256:8a703ca7604d42d74b2815eecf99a33359a8dccbb80806cce386d5e2dd992b05"},
    {file = "jschema_to_python-1.2.3.tar.gz", hash = "sha256:76ff14fe5d304708ccad1284e4b11f96a658949a31ee7faed9e0995279549b91"},
@@ -1427,6 +1478,7 @@ version = "2.0.0"
 description = "Diff JSON and JSON-like structures in Python"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "jsondiff-2.0.0-py3-none-any.whl", hash = "sha256:689841d66273fc88fc79f7d33f4c074774f4f214b6466e3aff0e5adaf889d1e0"},
    {file = "jsondiff-2.0.0.tar.gz", hash = "sha256:2795844ef075ec8a2b8d385c4d59f5ea48b08e7180fce3cb2787be0db00b1fb4"},
@@ -1438,6 +1490,8 @@ version = "0.20.0"
 description = "Python bindings for Jsonnet - The data templating language"
 optional = false
 python-versions = "*"
+groups = ["main"]
+markers = "python_version < \"3.13\""
 files = [
    {file = "jsonnet-0.20.0.tar.gz", hash = "sha256:7e770c7bf3a366b97b650a39430450f77612e74406731eb75c5bd59f3f104d4f"},
 ]
@@ -1448,6 +1502,7 @@ version = "1.32"
 description = "Apply JSON-Patches (RFC 6902)"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+groups = ["main"]
 files = [
    {file = "jsonpatch-1.32-py2.py3-none-any.whl", hash = "sha256:26ac385719ac9f54df8a2f0827bb8253aa3ea8ab7b3368457bcdb8c14595a397"},
    {file = "jsonpatch-1.32.tar.gz", hash = "sha256:b6ddfe6c3db30d81a96aaeceb6baf916094ffa23d7dd5fa2c13e13f8b6e600c2"},
@@ -1462,6 +1517,7 @@ version = "1.6.1"
 description = "A final implementation of JSONPath for Python that aims to be standard compliant, including arithmetic and binary comparison operators and providing clear AST for metaprogramming."
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "jsonpath-ng-1.6.1.tar.gz", hash = "sha256:086c37ba4917304850bd837aeab806670224d3f038fe2833ff593a672ef0a5fa"},
    {file = "jsonpath_ng-1.6.1-py3-none-any.whl", hash = "sha256:8f22cd8273d7772eea9aaa84d922e0841aa36fdb8a2c6b7f6c3791a16a9bc0be"},
@@ -1476,6 +1532,7 @@ version = "2.2.0"
 description = "Python library for serializing any arbitrary object graph into JSON"
 optional = false
 python-versions = ">=2.7"
+groups = ["main"]
 files = [
    {file = "jsonpickle-2.2.0-py2.py3-none-any.whl", hash = "sha256:de7f2613818aa4f234138ca11243d6359ff83ae528b2185efdd474f62bcf9ae1"},
    {file = "jsonpickle-2.2.0.tar.gz", hash = "sha256:7b272918b0554182e53dc340ddd62d9b7f902fec7e7b05620c04f3ccef479a0e"},
@@ -1492,6 +1549,7 @@ version = "2.3"
 description = "Identify specific nodes in a JSON document (RFC 6901)"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
+groups = ["main"]
 files = [
    {file = "jsonpointer-2.3-py2.py3-none-any.whl", hash = "sha256:51801e558539b4e9cd268638c078c6c5746c9ac96bc38152d443400e4f3793e9"},
    {file = "jsonpointer-2.3.tar.gz", hash = "sha256:97cba51526c829282218feb99dab1b1e6bdf8efd1c43dc9d57be093c0d69c99a"},
@@ -1503,6 +1561,7 @@ version = "4.17.3"
 description = "An implementation of JSON Schema validation for Python"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "jsonschema-4.17.3-py3-none-any.whl", hash = "sha256:a870ad254da1a8ca84b6a2905cac29d265f805acc57af304784962a2aa6508f6"},
    {file = "jsonschema-4.17.3.tar.gz", hash = "sha256:0f864437ab8b6076ba6707453ef8f98a6a0d512a80e93f8abdb676f737ecb60d"},
@@ -1522,6 +1581,7 @@ version = "0.1.6"
 description = "JSONSchema Spec with object-oriented paths"
 optional = false
 python-versions = ">=3.7.0,<4.0.0"
+groups = ["main"]
 files = [
    {file = "jsonschema_spec-0.1.6-py3-none-any.whl", hash = "sha256:f2206d18c89d1824c1f775ba14ed039743b41a9167bd2c5bdb774b66b3ca0bbf"},
    {file = "jsonschema_spec-0.1.6.tar.gz", hash = "sha256:90215863b56e212086641956b20127ccbf6d8a3a38343dad01d6a74d19482f76"},
@@ -1539,6 +1599,7 @@ version = "1.9"
 description = "Creates JUnit XML test result documents that can be read by tools such as Jenkins"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "junit-xml-1.9.tar.gz", hash = "sha256:de16a051990d4e25a3982b2dd9e89d671067548718866416faec14d9de56db9f"},
    {file = "junit_xml-1.9-py2.py3-none-any.whl", hash = "sha256:ec5ca1a55aefdd76d28fcc0b135251d156c7106fa979686a4b48d62b761b4732"},
@@ -1553,6 +1614,7 @@ version = "1.5.6"
 description = "Implementation of JOSE Web standards"
 optional = false
 python-versions = ">= 3.8"
+groups = ["main"]
 files = [
    {file = "jwcrypto-1.5.6-py3-none-any.whl", hash = "sha256:150d2b0ebbdb8f40b77f543fb44ffd2baeff48788be71f67f03566692fd55789"},
    {file = "jwcrypto-1.5.6.tar.gz", hash = "sha256:771a87762a0c081ae6166958a954f80848820b2ab066937dc8b8379d65b1b039"},
@@ -1568,6 +1630,7 @@ version = "2.0.2"
 description = "Pure Python client for Apache Kafka"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "kafka-python-2.0.2.tar.gz", hash = "sha256:04dfe7fea2b63726cd6f3e79a2d86e709d608d74406638c5da33a01d45a9d7e3"},
    {file = "kafka_python-2.0.2-py2.py3-none-any.whl", hash = "sha256:2d92418c7cb1c298fa6c7f0fb3519b520d0d7526ac6cb7ae2a4fc65a51a94b6e"},
@@ -1582,6 +1645,7 @@ version = "1.10.0"
 description = "A fast and thorough lazy object proxy."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "lazy-object-proxy-1.10.0.tar.gz", hash = "sha256:78247b6d45f43a52ef35c25b5581459e85117225408a4128a3daf8bf9648ac69"},
    {file = "lazy_object_proxy-1.10.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:855e068b0358ab916454464a884779c7ffa312b8925c6f7401e952dcf3b89977"},
@@ -1628,6 +1692,7 @@ version = "4.3.3"
 description = "LZ4 Bindings for Python"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "lz4-4.3.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b891880c187e96339474af2a3b2bfb11a8e4732ff5034be919aa9029484cd201"},
    {file = "lz4-4.3.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:222a7e35137d7539c9c33bb53fcbb26510c5748779364014235afc62b0ec797f"},
@@ -1678,6 +1743,7 @@ version = "2.1.1"
 description = "Safely add untrusted strings to HTML/XML markup."
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "MarkupSafe-2.1.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:86b1f75c4e7c2ac2ccdaec2b9022845dbb81880ca318bb7a0a01fbf7813e3812"},
    {file = "MarkupSafe-2.1.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:f121a1420d4e173a5d96e47e9a0c0dcff965afdf1626d28de1460815f7c4ee7a"},
@@ -1727,6 +1793,7 @@ version = "5.0.6"
 description = ""
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "moto-5.0.6-py2.py3-none-any.whl", hash = "sha256:ca1e22831a741733b581ff2ef4d6ae2e1c6db1eab97af1b78b86ca2c6e88c609"},
    {file = "moto-5.0.6.tar.gz", hash = "sha256:ad8b23f2b555ad694da8b2432a42b6d96beaaf67a4e7d932196a72193a2eee2c"},
@@ -1786,6 +1853,7 @@ version = "1.3.0"
 description = "Python library for arbitrary-precision floating-point arithmetic"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c"},
    {file = "mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f"},
@@ -1803,6 +1871,7 @@ version = "6.0.5"
 description = "multidict implementation"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "multidict-6.0.5-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:228b644ae063c10e7f324ab1ab6b548bdf6f8b47f3ec234fef1093bc2735e5f9"},
    {file = "multidict-6.0.5-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:896ebdcf62683551312c30e20614305f53125750803b614e9e6ce74a96232604"},
@@ -1902,6 +1971,7 @@ version = "1.13.0"
 description = "Optional static typing for Python"
 optional = false
 python-versions = ">=3.8"
+groups = ["dev"]
 files = [
    {file = "mypy-1.13.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:6607e0f1dd1fb7f0aca14d936d13fd19eba5e17e1cd2a14f808fa5f8f6d8f60a"},
    {file = "mypy-1.13.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:8a21be69bd26fa81b1f80a61ee7ab05b076c674d9b18fb56239d72e21d9f4c80"},
@@ -1954,6 +2024,7 @@ version = "1.26.0.post1"
 description = "Type annotations for boto3.S3 1.26.0 service generated with mypy-boto3-builder 7.11.10"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "mypy-boto3-s3-1.26.0.post1.tar.gz", hash = "sha256:6d7079f8c739dc993cbedad0736299c413b297814b73795a3855a79169ecc938"},
    {file = "mypy_boto3_s3-1.26.0.post1-py3-none-any.whl", hash = "sha256:7de2792ff0cc541b84cd46ff3a6aa2b6e5f267217f2203f27f6e4016bddc644d"},
@@ -1968,6 +2039,7 @@ version = "1.0.0"
 description = "Type system extensions for programs checked with the mypy type checker."
 optional = false
 python-versions = ">=3.5"
+groups = ["dev"]
 files = [
    {file = "mypy_extensions-1.0.0-py3-none-any.whl", hash = "sha256:4392f6c0eb8a5668a69e23d168ffa70f0be9ccfd32b5cc2d26a34ae5b844552d"},
    {file = "mypy_extensions-1.0.0.tar.gz", hash = "sha256:75dbf8955dc00442a438fc4d0666508a9a97b6bd41aa2f0ffe9d2f2725af0782"},
@@ -1979,6 +2051,7 @@ version = "2.8.5"
 description = "Python package for creating and manipulating graphs and networks"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "networkx-2.8.5-py3-none-any.whl", hash = "sha256:a762f4b385692d9c3a6f2912d058d76d29a827deaedf9e63ed14d397b8030687"},
    {file = "networkx-2.8.5.tar.gz", hash = "sha256:15a7b81a360791c458c55a417418ea136c13378cfdc06a2dcdc12bd2f9cf09c1"},
@@ -1997,6 +2070,7 @@ version = "0.4.4"
 description = "OpenAPI schema validation for Python"
 optional = false
 python-versions = ">=3.7.0,<4.0.0"
+groups = ["main"]
 files = [
    {file = "openapi_schema_validator-0.4.4-py3-none-any.whl", hash = "sha256:79f37f38ef9fd5206b924ed7a6f382cea7b649b3b56383c47f1906082b7b9015"},
    {file = "openapi_schema_validator-0.4.4.tar.gz", hash = "sha256:c573e2be2c783abae56c5a1486ab716ca96e09d1c3eab56020d1dc680aa57bf8"},
@@ -2015,6 +2089,7 @@ version = "0.5.7"
 description = "OpenAPI 2.0 (aka Swagger) and OpenAPI 3 spec validator"
 optional = false
 python-versions = ">=3.7.0,<4.0.0"
+groups = ["main"]
 files = [
    {file = "openapi_spec_validator-0.5.7-py3-none-any.whl", hash = "sha256:8712d2879db7692974ef89c47a3ebfc79436442921ec3a826ac0ce80cde8c549"},
    {file = "openapi_spec_validator-0.5.7.tar.gz", hash = "sha256:6c2d42180045a80fd6314de848b94310bdb0fa4949f4b099578b69f79d9fa5ac"},
@@ -2032,6 +2107,7 @@ version = "24.2"
 description = "Core utilities for Python packages"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "packaging-24.2-py3-none-any.whl", hash = "sha256:09abb1bccd265c01f4a3aa3f7a7db064b36514d2cba19a2f694fe6150451a759"},
    {file = "packaging-24.2.tar.gz", hash = "sha256:c228a6dc5e932d346bc5739379109d49e8853dd8223571c7c5b55260edc0b97f"},
@@ -2043,6 +2119,7 @@ version = "0.4.3"
 description = "Object-oriented paths"
 optional = false
 python-versions = ">=3.7.0,<4.0.0"
+groups = ["main"]
 files = [
    {file = "pathable-0.4.3-py3-none-any.whl", hash = "sha256:cdd7b1f9d7d5c8b8d3315dbf5a86b2596053ae845f056f57d97c0eefff84da14"},
    {file = "pathable-0.4.3.tar.gz", hash = "sha256:5c869d315be50776cc8a993f3af43e0c60dc01506b399643f919034ebf4cdcab"},
@@ -2054,6 +2131,7 @@ version = "5.9.0"
 description = "Python Build Reasonableness"
 optional = false
 python-versions = ">=2.6"
+groups = ["main"]
 files = [
    {file = "pbr-5.9.0-py2.py3-none-any.whl", hash = "sha256:e547125940bcc052856ded43be8e101f63828c2d94239ffbe2b327ba3d5ccf0a"},
    {file = "pbr-5.9.0.tar.gz", hash = "sha256:e8dca2f4b43560edef58813969f52a56cef023146cbb8931626db80e6c1c4308"},
@@ -2065,6 +2143,7 @@ version = "1.0.0"
 description = "plugin and hook calling mechanisms for python"
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "pluggy-1.0.0-py2.py3-none-any.whl", hash = "sha256:74134bbf457f031a36d68416e1509f34bd5ccc019f0bcc952c7b909d06b37bd3"},
    {file = "pluggy-1.0.0.tar.gz", hash = "sha256:4224373bacce55f955a878bf9cfa763c1e360858e330072059e10bad68531159"},
@@ -2080,6 +2159,7 @@ version = "3.11"
 description = "Python Lex & Yacc"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "ply-3.11-py2.py3-none-any.whl", hash = "sha256:096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce"},
    {file = "ply-3.11.tar.gz", hash = "sha256:00c7c1aaa88358b9c765b6d3000c6eec0ba42abca5351b095321aef446081da3"},
@@ -2091,6 +2171,7 @@ version = "0.14.1"
 description = "Python client for the Prometheus monitoring system."
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "prometheus_client-0.14.1-py3-none-any.whl", hash = "sha256:522fded625282822a89e2773452f42df14b5a8e84a86433e3f8a189c1d54dc01"},
    {file = "prometheus_client-0.14.1.tar.gz", hash = "sha256:5459c427624961076277fdc6dc50540e2bacb98eebde99886e59ec55ed92093a"},
@@ -2105,6 +2186,7 @@ version = "0.2.0"
 description = "Accelerated property cache"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "propcache-0.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:c5869b8fd70b81835a6f187c5fdbe67917a04d7e52b6e7cc4e5fe39d55c39d58"},
    {file = "propcache-0.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:952e0d9d07609d9c5be361f33b0d6d650cd2bae393aabb11d9b719364521984b"},
@@ -2212,6 +2294,7 @@ version = "5.9.4"
 description = "Cross-platform lib for process and system monitoring in Python."
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
+groups = ["main"]
 files = [
    {file = "psutil-5.9.4-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:c1ca331af862803a42677c120aff8a814a804e09832f166f226bfd22b56feee8"},
    {file = "psutil-5.9.4-cp27-cp27m-manylinux2010_i686.whl", hash = "sha256:68908971daf802203f3d37e78d3f8831b6d1014864d7a85937941bb35f09aefe"},
@@ -2238,6 +2321,7 @@ version = "2.9.10"
 description = "psycopg2 - Python-PostgreSQL Database Adapter"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "psycopg2-binary-2.9.10.tar.gz", hash = "sha256:4b3df0e6990aa98acda57d983942eff13d824135fe2250e6522edaa782a06de2"},
    {file = "psycopg2_binary-2.9.10-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:0ea8e3d0ae83564f2fc554955d327fa081d065c8ca5cc6d2abb643e2c9c1200f"},
@@ -2286,6 +2370,7 @@ files = [
    {file = "psycopg2_binary-2.9.10-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:bb89f0a835bcfc1d42ccd5f41f04870c1b936d8507c6df12b7737febc40f0909"},
    {file = "psycopg2_binary-2.9.10-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:f0c2d907a1e102526dd2986df638343388b94c33860ff3bbe1384130828714b1"},
    {file = "psycopg2_binary-2.9.10-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f8157bed2f51db683f31306aa497311b560f2265998122abe1dce6428bd86567"},
+    {file = "psycopg2_binary-2.9.10-cp313-cp313-win_amd64.whl", hash = "sha256:27422aa5f11fbcd9b18da48373eb67081243662f9b46e6fd07c3eb46e4535142"},
    {file = "psycopg2_binary-2.9.10-cp38-cp38-macosx_12_0_x86_64.whl", hash = "sha256:eb09aa7f9cecb45027683bb55aebaaf45a0df8bf6de68801a6afdc7947bb09d4"},
    {file = "psycopg2_binary-2.9.10-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b73d6d7f0ccdad7bc43e6d34273f70d587ef62f824d7261c4ae9b8b1b6af90e8"},
    {file = "psycopg2_binary-2.9.10-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ce5ab4bf46a211a8e924d307c1b1fcda82368586a19d0a24f8ae166f5c784864"},
@@ -2314,6 +2399,7 @@ version = "0.5.4"
 description = "Pure Python PartiQL Parser"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "py_partiql_parser-0.5.4-py2.py3-none-any.whl", hash = "sha256:3dc4295a47da9587681a96b35c6e151886fdbd0a4acbe0d97c4c68e5f689d315"},
    {file = "py_partiql_parser-0.5.4.tar.gz", hash = "sha256:72e043919538fa63edae72fb59afc7e3fd93adbde656718a7d2b4666f23dd114"},
@@ -2328,6 +2414,7 @@ version = "2.21"
 description = "C parser in Python"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
+groups = ["main"]
 files = [
    {file = "pycparser-2.21-py2.py3-none-any.whl", hash = "sha256:8ee45429555515e1f6b185e78100aea234072576aa43ab53aefcae078162fca9"},
    {file = "pycparser-2.21.tar.gz", hash = "sha256:e644fdec12f7872f86c58ff790da456218b10f863970249516d60a5eaca77206"},
@@ -2339,6 +2426,7 @@ version = "2.10.4"
 description = "Data validation using Python type hints"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "pydantic-2.10.4-py3-none-any.whl", hash = "sha256:597e135ea68be3a37552fb524bc7d0d66dcf93d395acd93a00682f1efcb8ee3d"},
    {file = "pydantic-2.10.4.tar.gz", hash = "sha256:82f12e9723da6de4fe2ba888b5971157b3be7ad914267dea8f05f82b28254f06"},
@@ -2359,6 +2447,7 @@ version = "2.27.2"
 description = "Core functionality for Pydantic validation and serialization"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "pydantic_core-2.27.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:2d367ca20b2f14095a8f4fa1210f5a7b78b8a20009ecced6b12818f455b1e9fa"},
    {file = "pydantic_core-2.27.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:491a2b73db93fab69731eaee494f320faa4e093dbed776be1a829c2eb222c34c"},
@@ -2471,6 +2560,7 @@ version = "2.4.0"
 description = "JSON Web Token implementation in Python"
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "PyJWT-2.4.0-py3-none-any.whl", hash = "sha256:72d1d253f32dbd4f5c88eaf1fdc62f3a19f676ccbadb9dbc5d07e951b2b26daf"},
    {file = "PyJWT-2.4.0.tar.gz", hash = "sha256:d42908208c699b3b973cbeb01a969ba6a96c821eefb1c5bfe4c390c01d67abba"},
@@ -2491,6 +2581,7 @@ version = "3.0.9"
 description = "pyparsing module - Classes and methods to define and execute parsing grammars"
 optional = false
 python-versions = ">=3.6.8"
+groups = ["main"]
 files = [
    {file = "pyparsing-3.0.9-py3-none-any.whl", hash = "sha256:5026bae9a10eeaefb61dab2f09052b9f4307d44aee4eda64b309723d8d206bbc"},
    {file = "pyparsing-3.0.9.tar.gz", hash = "sha256:2b020ecf7d21b687f219b71ecad3631f644a47f01403fa1d1036b0c6416d70fb"},
@@ -2505,6 +2596,7 @@ version = "0.18.1"
 description = "Persistent/Functional/Immutable data structures"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "pyrsistent-0.18.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:df46c854f490f81210870e509818b729db4488e1f30f2a1ce1698b2295a878d1"},
    {file = "pyrsistent-0.18.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5d45866ececf4a5fff8742c25722da6d4c9e180daa7b405dc0a2a2790d668c26"},
@@ -2535,6 +2627,7 @@ version = "7.4.4"
 description = "pytest: simple powerful testing with Python"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "pytest-7.4.4-py3-none-any.whl", hash = "sha256:b090cdf5ed60bf4c45261be03239c2c1c22df034fbffe691abe93cd80cea01d8"},
    {file = "pytest-7.4.4.tar.gz", hash = "sha256:2cf0005922c6ace4a3e2ec8b4080eb0d9753fdc93107415332f50ce9e7994280"},
@@ -2555,6 +2648,7 @@ version = "0.21.0"
 description = "Pytest support for asyncio"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "pytest-asyncio-0.21.0.tar.gz", hash = "sha256:2b38a496aef56f56b0e87557ec313e11e1ab9276fc3863f6a7be0f1d0e415e1b"},
    {file = "pytest_asyncio-0.21.0-py3-none-any.whl", hash = "sha256:f2b3366b7cd501a4056858bd39349d5af19742aed2d81660b7998b6341c7eb9c"},
@@ -2573,6 +2667,7 @@ version = "1.0.8"
 description = "pytest-httpserver is a httpserver for pytest"
 optional = false
 python-versions = ">=3.8,<4.0"
+groups = ["main"]
 files = [
    {file = "pytest_httpserver-1.0.8-py3-none-any.whl", hash = "sha256:24cd3d9f6a0b927c7bfc400d0b3fda7442721b8267ce29942bf307b190f0bb09"},
    {file = "pytest_httpserver-1.0.8.tar.gz", hash = "sha256:e052f69bc8a9073db02484681e8e47004dd1fb3763b0ae833bd899e5895c559a"},
@@ -2587,6 +2682,7 @@ version = "0.6.3"
 description = "It helps to use fixtures in pytest.mark.parametrize"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "pytest-lazy-fixture-0.6.3.tar.gz", hash = "sha256:0e7d0c7f74ba33e6e80905e9bfd81f9d15ef9a790de97993e34213deb5ad10ac"},
    {file = "pytest_lazy_fixture-0.6.3-py3-none-any.whl", hash = "sha256:e0b379f38299ff27a653f03eaa69b08a6fd4484e46fd1c9907d984b9f9daeda6"},
@@ -2601,6 +2697,7 @@ version = "1.1.0"
 description = "pytest plugin to run your tests in a specific order"
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "pytest-order-1.1.0.tar.gz", hash = "sha256:139d25b30826b78eebb42722f747eab14c44b88059d7a71d4f79d14a057269a5"},
    {file = "pytest_order-1.1.0-py3-none-any.whl", hash = "sha256:3b3730969c97900fa5cd31ecff80847680ed56b2490954565c14949ba60d9371"},
@@ -2615,6 +2712,7 @@ version = "0.9.3"
 description = "pytest plugin for repeating tests"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "pytest_repeat-0.9.3-py3-none-any.whl", hash = "sha256:26ab2df18226af9d5ce441c858f273121e92ff55f5bb311d25755b8d7abdd8ed"},
    {file = "pytest_repeat-0.9.3.tar.gz", hash = "sha256:ffd3836dfcd67bb270bec648b330e20be37d2966448c4148c4092d1e8aba8185"},
@@ -2629,6 +2727,7 @@ version = "15.0"
 description = "pytest plugin to re-run tests to eliminate flaky failures"
 optional = false
 python-versions = ">=3.9"
+groups = ["main"]
 files = [
    {file = "pytest-rerunfailures-15.0.tar.gz", hash = "sha256:2d9ac7baf59f4c13ac730b47f6fa80e755d1ba0581da45ce30b72fb3542b4474"},
    {file = "pytest_rerunfailures-15.0-py3-none-any.whl", hash = "sha256:dd150c4795c229ef44320adc9a0c0532c51b78bb7a6843a8c53556b9a611df1a"},
@@ -2644,6 +2743,7 @@ version = "0.8.1"
 description = "Pytest plugin which splits the test suite to equally sized sub suites based on test execution time."
 optional = false
 python-versions = ">=3.7.1,<4.0"
+groups = ["main"]
 files = [
    {file = "pytest_split-0.8.1-py3-none-any.whl", hash = "sha256:74b110ea091bd147cc1c5f9665a59506e5cedfa66f96a89fb03e4ab447c2c168"},
    {file = "pytest_split-0.8.1.tar.gz", hash = "sha256:2d88bd3dc528689a7a3f58fc12ea165c3aa62e90795e420dfad920afe5612d6d"},
@@ -2658,6 +2758,7 @@ version = "2.1.0"
 description = "pytest plugin to abort hanging tests"
 optional = false
 python-versions = ">=3.6"
+groups = ["main"]
 files = [
    {file = "pytest-timeout-2.1.0.tar.gz", hash = "sha256:c07ca07404c612f8abbe22294b23c368e2e5104b521c1790195561f37e1ac3d9"},
    {file = "pytest_timeout-2.1.0-py3-none-any.whl", hash = "sha256:f6f50101443ce70ad325ceb4473c4255e9d74e3c7cd0ef827309dfa4c0d975c6"},
@@ -2672,6 +2773,7 @@ version = "3.3.1"
 description = "pytest xdist plugin for distributed testing, most importantly across multiple CPUs"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "pytest-xdist-3.3.1.tar.gz", hash = "sha256:d5ee0520eb1b7bcca50a60a518ab7a7707992812c578198f8b44fdfac78e8c93"},
    {file = "pytest_xdist-3.3.1-py3-none-any.whl", hash = "sha256:ff9daa7793569e6a68544850fd3927cd257cc03a7ef76c95e86915355e82b5f2"},
@@ -2692,6 +2794,7 @@ version = "2.8.2"
 description = "Extensions to the standard Python datetime module"
 optional = false
 python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,>=2.7"
+groups = ["main"]
 files = [
    {file = "python-dateutil-2.8.2.tar.gz", hash = "sha256:0123cacc1627ae19ddf3c27a5de5bd67ee4586fbdd6440d9748f8abb483d3e86"},
    {file = "python_dateutil-2.8.2-py2.py3-none-any.whl", hash = "sha256:961d03dc3453ebbc59dbdea9e4e11c5651520a876d0f4db161e8674aae935da9"},
@@ -2706,6 +2809,7 @@ version = "1.0.1"
 description = "Read key-value pairs from a .env file and set them as environment variables"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "python-dotenv-1.0.1.tar.gz", hash = "sha256:e324ee90a023d808f1959c46bcbc04446a10ced277783dc6ee09987c37ec10ca"},
    {file = "python_dotenv-1.0.1-py3-none-any.whl", hash = "sha256:f7b63ef50f1b690dddf550d03497b66d609393b40b564ed0d674909a68ebf16a"},
@@ -2720,6 +2824,7 @@ version = "2024.1"
 description = "World timezone definitions, modern and historical"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "pytz-2024.1-py2.py3-none-any.whl", hash = "sha256:328171f4e3623139da4983451950b28e95ac706e13f3f2630a879749e7a8b319"},
    {file = "pytz-2024.1.tar.gz", hash = "sha256:2a29735ea9c18baf14b448846bde5a48030ed267578472d8955cd0e7443a9812"},
@@ -2731,6 +2836,8 @@ version = "308"
 description = "Python for Window Extensions"
 optional = false
 python-versions = "*"
+groups = ["main"]
+markers = "sys_platform == \"win32\""
 files = [
    {file = "pywin32-308-cp310-cp310-win32.whl", hash = "sha256:796ff4426437896550d2981b9c2ac0ffd75238ad9ea2d3bfa67a1abd546d262e"},
    {file = "pywin32-308-cp310-cp310-win_amd64.whl", hash = "sha256:4fc888c59b3c0bef905ce7eb7e2106a07712015ea1c8234b703a088d46110e8e"},
@@ -2758,6 +2865,7 @@ version = "6.0.2"
 description = "YAML parser and emitter for Python"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "PyYAML-6.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0a9a2848a5b7feac301353437eb7d5957887edbf81d56e903999a75a3d743086"},
    {file = "PyYAML-6.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:29717114e51c84ddfba879543fb232a6ed60086602313ca38cce623c1d62cfbf"},
@@ -2820,6 +2928,7 @@ version = "2024.4.28"
 description = "Alternative regular expression module, to replace re."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "regex-2024.4.28-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:cd196d056b40af073d95a2879678585f0b74ad35190fac04ca67954c582c6b61"},
    {file = "regex-2024.4.28-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:8bb381f777351bd534462f63e1c6afb10a7caa9fa2a421ae22c26e796fe31b1f"},
@@ -2908,6 +3017,7 @@ version = "2.32.3"
 description = "Python HTTP for Humans."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "requests-2.32.3-py3-none-any.whl", hash = "sha256:70761cfe03c773ceb22aa2f671b4757976145175cdfca038c02654d061d6dcc6"},
    {file = "requests-2.32.3.tar.gz", hash = "sha256:55365417734eb18255590a9ff9eb97e9e1da868d4ccd6402399eaf68af20a760"},
@@ -2929,6 +3039,7 @@ version = "0.25.3"
 description = "A utility library for mocking out the `requests` Python library."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "responses-0.25.3-py3-none-any.whl", hash = "sha256:521efcbc82081ab8daa588e08f7e8a64ce79b91c39f6e62199b19159bea7dbcb"},
    {file = "responses-0.25.3.tar.gz", hash = "sha256:617b9247abd9ae28313d57a75880422d55ec63c29d33d629697590a034358dba"},
@@ -2948,6 +3059,7 @@ version = "0.1.4"
 description = "A pure python RFC3339 validator"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+groups = ["main"]
 files = [
    {file = "rfc3339_validator-0.1.4-py2.py3-none-any.whl", hash = "sha256:24f6ec1eda14ef823da9e36ec7113124b39c04d50a4d3d3a3c2859577e7791fa"},
    {file = "rfc3339_validator-0.1.4.tar.gz", hash = "sha256:138a2abdf93304ad60530167e51d2dfb9549521a836871b88d7f4695d0022f6b"},
@@ -2962,6 +3074,7 @@ version = "0.7.0"
 description = "An extremely fast Python linter and code formatter, written in Rust."
 optional = false
 python-versions = ">=3.7"
+groups = ["dev"]
 files = [
    {file = "ruff-0.7.0-py3-none-linux_armv6l.whl", hash = "sha256:0cdf20c2b6ff98e37df47b2b0bd3a34aaa155f59a11182c1303cce79be715628"},
    {file = "ruff-0.7.0-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:496494d350c7fdeb36ca4ef1c9f21d80d182423718782222c29b3e72b3512737"},
@@ -2989,6 +3102,7 @@ version = "0.10.0"
 description = "An Amazon S3 Transfer Manager"
 optional = false
 python-versions = ">= 3.8"
+groups = ["main"]
 files = [
    {file = "s3transfer-0.10.0-py3-none-any.whl", hash = "sha256:3cdb40f5cfa6966e812209d0994f2a4709b561c88e90cf00c2696d2df4e56b2e"},
    {file = "s3transfer-0.10.0.tar.gz", hash = "sha256:d0c8bbf672d5eebbe4e57945e23b972d963f07d82f661cabf678a5c88831595b"},
@@ -3006,6 +3120,7 @@ version = "1.0.4"
 description = "Classes implementing the SARIF 2.1.0 object model."
 optional = false
 python-versions = ">= 2.7"
+groups = ["main"]
 files = [
    {file = "sarif_om-1.0.4-py3-none-any.whl", hash = "sha256:539ef47a662329b1c8502388ad92457425e95dc0aaaf995fe46f4984c4771911"},
    {file = "sarif_om-1.0.4.tar.gz", hash = "sha256:cd5f416b3083e00d402a92e449a7ff67af46f11241073eea0461802a3b5aef98"},
@@ -3021,6 +3136,7 @@ version = "70.0.0"
 description = "Easily download, build, install, upgrade, and uninstall Python packages"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "setuptools-70.0.0-py3-none-any.whl", hash = "sha256:54faa7f2e8d2d11bcd2c07bed282eef1046b5c080d1c32add737d7b5817b1ad4"},
    {file = "setuptools-70.0.0.tar.gz", hash = "sha256:f211a66637b8fa059bb28183da127d4e86396c991a942b028c6650d4319c3fd0"},
@@ -3036,6 +3152,7 @@ version = "1.16.0"
 description = "Python 2 and 3 compatibility utilities"
 optional = false
 python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*"
+groups = ["main"]
 files = [
    {file = "six-1.16.0-py2.py3-none-any.whl", hash = "sha256:8abb2f1d86890a2dfb989f9a77cfcfd3e47c2a354b01111771326f8aa26e0254"},
    {file = "six-1.16.0.tar.gz", hash = "sha256:1e61c37477a1626458e36f7b1d82aa5c9b094fa4802892072e49de9c60c4c926"},
@@ -3047,6 +3164,7 @@ version = "1.3.0"
 description = "Sniff out which async library your code is running under"
 optional = false
 python-versions = ">=3.7"
+groups = ["main"]
 files = [
    {file = "sniffio-1.3.0-py3-none-any.whl", hash = "sha256:eecefdce1e5bbfb7ad2eeaabf7c1eeb404d7757c379bd1f7e5cce9d8bf425384"},
    {file = "sniffio-1.3.0.tar.gz", hash = "sha256:e60305c5e5d314f5389259b7f22aaa33d8f7dee49763119234af3755c55b9101"},
@@ -3058,6 +3176,7 @@ version = "1.12"
 description = "Computer algebra system (CAS) in Python"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "sympy-1.12-py3-none-any.whl", hash = "sha256:c3588cd4295d0c0f603d0f2ae780587e64e2efeedb3521e46b9bb1d08d184fa5"},
    {file = "sympy-1.12.tar.gz", hash = "sha256:ebf595c8dac3e0fdc4152c51878b498396ec7f30e7a914d6071e674d49420fb8"},
@@ -3072,6 +3191,7 @@ version = "4.9.0"
 description = "Python library for throwaway instances of anything that can run in a Docker container"
 optional = false
 python-versions = "<4.0,>=3.9"
+groups = ["main"]
 files = [
    {file = "testcontainers-4.9.0-py3-none-any.whl", hash = "sha256:c6fee929990972c40bf6b91b7072c94064ff3649b405a14fde0274c8b2479d32"},
    {file = "testcontainers-4.9.0.tar.gz", hash = "sha256:2cd6af070109ff68c1ab5389dc89c86c2dc3ab30a21ca734b2cb8f0f80ad479e"},
@@ -3125,6 +3245,7 @@ version = "0.10.2"
 description = "Python Library for Tom's Obvious, Minimal Language"
 optional = false
 python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*"
+groups = ["main"]
 files = [
    {file = "toml-0.10.2-py2.py3-none-any.whl", hash = "sha256:806143ae5bfb6a3c6e736a764057db0e6a0e05e338b5630894a5f779cabb4f9b"},
    {file = "toml-0.10.2.tar.gz", hash = "sha256:b3bda1d108d5dd99f4a20d24d9c348e91c4db7ab1b749200bded2f839ccbe68f"},
@@ -3136,6 +3257,7 @@ version = "1.5.0.20240925"
 description = "Typing stubs for jwcrypto"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "types-jwcrypto-1.5.0.20240925.tar.gz", hash = "sha256:50e17b790378c96239344476c7bd13b52d0c7eeb6d16c2d53723e48cc6bbf4fe"},
    {file = "types_jwcrypto-1.5.0.20240925-py3-none-any.whl", hash = "sha256:2d12a2d528240d326075e896aafec7056b9136bf3207fa6ccf3fcb8fbf9e11a1"},
@@ -3150,6 +3272,7 @@ version = "5.9.5.12"
 description = "Typing stubs for psutil"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "types-psutil-5.9.5.12.tar.gz", hash = "sha256:61a91679d3fe737250013b624dca09375e7cc3ad77dcc734553746c429c02aca"},
    {file = "types_psutil-5.9.5.12-py3-none-any.whl", hash = "sha256:e9a147b8561235c6afcce5aa1adb973fad9ab2c50cf89820697687f53510358f"},
@@ -3161,6 +3284,7 @@ version = "2.9.21.20241019"
 description = "Typing stubs for psycopg2"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "types-psycopg2-2.9.21.20241019.tar.gz", hash = "sha256:bca89b988d2ebd19bcd08b177d22a877ea8b841decb10ed130afcf39404612fa"},
    {file = "types_psycopg2-2.9.21.20241019-py3-none-any.whl", hash = "sha256:44d091e67732d16a941baae48cd7b53bf91911bc36888652447cf1ef0c1fb3f6"},
@@ -3172,6 +3296,7 @@ version = "0.6.3.3"
 description = "Typing stubs for pytest-lazy-fixture"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "types-pytest-lazy-fixture-0.6.3.3.tar.gz", hash = "sha256:2ef79d66bcde0e50acdac8dc55074b9ae0d4cfaeabdd638f5522f4cac7c8a2c7"},
    {file = "types_pytest_lazy_fixture-0.6.3.3-py3-none-any.whl", hash = "sha256:a56a55649147ff960ff79d4b2c781a4f769351abc1876873f3116d0bd0c96353"},
@@ -3183,6 +3308,7 @@ version = "6.0.12.20240917"
 description = "Typing stubs for PyYAML"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "types-PyYAML-6.0.12.20240917.tar.gz", hash = "sha256:d1405a86f9576682234ef83bcb4e6fff7c9305c8b1fbad5e0bcd4f7dbdc9c587"},
    {file = "types_PyYAML-6.0.12.20240917-py3-none-any.whl", hash = "sha256:392b267f1c0fe6022952462bf5d6523f31e37f6cea49b14cee7ad634b6301570"},
@@ -3194,6 +3320,7 @@ version = "2.31.0.0"
 description = "Typing stubs for requests"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "types-requests-2.31.0.0.tar.gz", hash = "sha256:c1c29d20ab8d84dff468d7febfe8e0cb0b4664543221b386605e14672b44ea25"},
    {file = "types_requests-2.31.0.0-py3-none-any.whl", hash = "sha256:7c5cea7940f8e92ec560bbc468f65bf684aa3dcf0554a6f8c4710f5f708dc598"},
@@ -3208,6 +3335,7 @@ version = "0.6.0.post3"
 description = "Type annotations and code completion for s3transfer"
 optional = false
 python-versions = ">=3.7,<4.0"
+groups = ["main"]
 files = [
    {file = "types-s3transfer-0.6.0.post3.tar.gz", hash = "sha256:92c3704e5d041202bfb5ddb79d083fd1a02de2c5dfec6a91576823e6b5c93993"},
    {file = "types_s3transfer-0.6.0.post3-py3-none-any.whl", hash = "sha256:eedc5117275565b3c83662c0ccc81662a34da5dda8bd502b89d296b6d5cb091d"},
@@ -3219,6 +3347,7 @@ version = "0.10.8.6"
 description = "Typing stubs for toml"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "types-toml-0.10.8.6.tar.gz", hash = "sha256:6d3ac79e36c9ee593c5d4fb33a50cca0e3adceb6ef5cff8b8e5aef67b4c4aaf2"},
    {file = "types_toml-0.10.8.6-py3-none-any.whl", hash = "sha256:de7b2bb1831d6f7a4b554671ffe5875e729753496961b3e9b202745e4955dafa"},
@@ -3230,6 +3359,7 @@ version = "1.26.17"
 description = "Typing stubs for urllib3"
 optional = false
 python-versions = "*"
+groups = ["main"]
 files = [
    {file = "types-urllib3-1.26.17.tar.gz", hash = "sha256:73fd274524c3fc7cd8cd9ceb0cb67ed99b45f9cb2831013e46d50c1451044800"},
    {file = "types_urllib3-1.26.17-py3-none-any.whl", hash = "sha256:0d027fcd27dbb3cb532453b4d977e05bc1e13aefd70519866af211b3003d895d"},
@@ -3241,6 +3371,7 @@ version = "4.12.2"
 description = "Backported and Experimental Type Hints for Python 3.8+"
 optional = false
 python-versions = ">=3.8"
+groups = ["main", "dev"]
 files = [
    {file = "typing_extensions-4.12.2-py3-none-any.whl", hash = "sha256:04e5ca0351e0f3f85c6853954072df659d0d13fac324d0072316b67d7794700d"},
    {file = "typing_extensions-4.12.2.tar.gz", hash = "sha256:1a7ead55c7e559dd4dee8856e3a88b41225abfe1ce8df57b7c13915fe121ffb8"},
@@ -3252,6 +3383,7 @@ version = "1.26.19"
 description = "HTTP library with thread-safe connection pooling, file post, and more."
 optional = false
 python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*,>=2.7"
+groups = ["main"]
 files = [
    {file = "urllib3-1.26.19-py2.py3-none-any.whl", hash = "sha256:37a0344459b199fce0e80b0d3569837ec6b6937435c5244e7fd73fa6006830f3"},
    {file = "urllib3-1.26.19.tar.gz", hash = "sha256:3e3d753a8618b86d7de333b4223005f68720bcd6a7d2bcb9fbd2229ec7c1e429"},
@@ -3268,6 +3400,7 @@ version = "12.0"
 description = "An implementation of the WebSocket Protocol (RFC 6455 & 7692)"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "websockets-12.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d554236b2a2006e0ce16315c16eaa0d628dab009c33b63ea03f41c6107958374"},
    {file = "websockets-12.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2d225bb6886591b1746b17c0573e29804619c8f755b5598d875bb4235ea639be"},
@@ -3349,6 +3482,7 @@ version = "3.0.6"
 description = "The comprehensive WSGI web application library."
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "werkzeug-3.0.6-py3-none-any.whl", hash = "sha256:1bc0c2310d2fbb07b1dd1105eba2f7af72f322e1e455f2f93c993bee8c8a5f17"},
    {file = "werkzeug-3.0.6.tar.gz", hash = "sha256:a8dd59d4de28ca70471a34cba79bed5f7ef2e036a76b3ab0835474246eb41f8d"},
@@ -3366,6 +3500,7 @@ version = "1.14.1"
 description = "Module for decorators, wrappers and monkey patching."
 optional = false
 python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,>=2.7"
+groups = ["main"]
 files = [
    {file = "wrapt-1.14.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:1b376b3f4896e7930f1f772ac4b064ac12598d1c38d04907e696cc4d794b43d3"},
    {file = "wrapt-1.14.1-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:903500616422a40a98a5a3c4ff4ed9d0066f3b4c951fa286018ecdf0750194ef"},
@@ -3386,6 +3521,16 @@ files = [
    {file = "wrapt-1.14.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:8ad85f7f4e20964db4daadcab70b47ab05c7c1cf2a7c1e51087bfaa83831854c"},
    {file = "wrapt-1.14.1-cp310-cp310-win32.whl", hash = "sha256:a9a52172be0b5aae932bef82a79ec0a0ce87288c7d132946d645eba03f0ad8a8"},
    {file = "wrapt-1.14.1-cp310-cp310-win_amd64.whl", hash = "sha256:6d323e1554b3d22cfc03cd3243b5bb815a51f5249fdcbb86fda4bf62bab9e164"},
+    {file = "wrapt-1.14.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ecee4132c6cd2ce5308e21672015ddfed1ff975ad0ac8d27168ea82e71413f55"},
+    {file = "wrapt-1.14.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2020f391008ef874c6d9e208b24f28e31bcb85ccff4f335f15a3251d222b92d9"},
+    {file = "wrapt-1.14.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2feecf86e1f7a86517cab34ae6c2f081fd2d0dac860cb0c0ded96d799d20b335"},
+    {file = "wrapt-1.14.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:240b1686f38ae665d1b15475966fe0472f78e71b1b4903c143a842659c8e4cb9"},
+    {file = "wrapt-1.14.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a9008dad07d71f68487c91e96579c8567c98ca4c3881b9b113bc7b33e9fd78b8"},
+    {file = "wrapt-1.14.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:6447e9f3ba72f8e2b985a1da758767698efa72723d5b59accefd716e9e8272bf"},
+    {file = "wrapt-1.14.1-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:acae32e13a4153809db37405f5eba5bac5fbe2e2ba61ab227926a22901051c0a"},
+    {file = "wrapt-1.14.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:49ef582b7a1152ae2766557f0550a9fcbf7bbd76f43fbdc94dd3bf07cc7168be"},
+    {file = "wrapt-1.14.1-cp311-cp311-win32.whl", hash = "sha256:358fe87cc899c6bb0ddc185bf3dbfa4ba646f05b1b0b9b5a27c2cb92c2cea204"},
+    {file = "wrapt-1.14.1-cp311-cp311-win_amd64.whl", hash = "sha256:26046cd03936ae745a502abf44dac702a5e6880b2b01c29aea8ddf3353b68224"},
    {file = "wrapt-1.14.1-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:43ca3bbbe97af00f49efb06e352eae40434ca9d915906f77def219b88e85d907"},
    {file = "wrapt-1.14.1-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:6b1a564e6cb69922c7fe3a678b9f9a3c54e72b469875aa8018f18b4d1dd1adf3"},
    {file = "wrapt-1.14.1-cp35-cp35m-manylinux2010_i686.whl", hash = "sha256:00b6d4ea20a906c0ca56d84f93065b398ab74b927a7a3dbd470f6fc503f95dc3"},
@@ -3439,6 +3584,7 @@ version = "0.13.0"
 description = "Makes working with XML feel like you are working with JSON"
 optional = false
 python-versions = ">=3.4"
+groups = ["main"]
 files = [
    {file = "xmltodict-0.13.0-py2.py3-none-any.whl", hash = "sha256:aa89e8fd76320154a40d19a0df04a4695fb9dc5ba977cbb68ab3e4eb225e7852"},
    {file = "xmltodict-0.13.0.tar.gz", hash = "sha256:341595a488e3e01a85a9d8911d8912fd922ede5fecc4dce437eb4b6c8d037e56"},
@@ -3450,6 +3596,7 @@ version = "1.17.2"
 description = "Yet another URL library"
 optional = false
 python-versions = ">=3.9"
+groups = ["main"]
 files = [
    {file = "yarl-1.17.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:93771146ef048b34201bfa382c2bf74c524980870bb278e6df515efaf93699ff"},
    {file = "yarl-1.17.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:8281db240a1616af2f9c5f71d355057e73a1409c4648c8949901396dc0a3c151"},
@@ -3546,6 +3693,7 @@ version = "0.23.0"
 description = "Zstandard bindings for Python"
 optional = false
 python-versions = ">=3.8"
+groups = ["main"]
 files = [
    {file = "zstandard-0.23.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bf0a05b6059c0528477fba9054d09179beb63744355cab9f38059548fedd46a9"},
    {file = "zstandard-0.23.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:fc9ca1c9718cb3b06634c7c8dec57d24e9438b2aa9a0f02b8bb36bf478538880"},
@@ -3653,6 +3801,6 @@ cffi = {version = ">=1.11", markers = "platform_python_implementation == \"PyPy\
 cffi = ["cffi (>=1.11)"]

 [metadata]
-lock-version = "2.0"
+lock-version = "2.1"
 python-versions = "^3.11"
-content-hash = "e6904aca09abc6c805604b21a5702a97e0056406f9ec7469b091d35ee10a6b16"
+content-hash = "4dc3165fe22c0e0f7a030ea0f8a680ae2ff74561d8658c393abbe9112caaf5d7"
--- a/proxy/src/context/parquet.rs
+++ b/proxy/src/context/parquet.rs
@@ -423,11 +423,11 @@ async fn upload_parquet(
    .await
    .ok_or_else(|| anyhow::Error::new(TimeoutOrCancel::Cancel))
    .and_then(|x| x)
-    .context("request_data_upload")
+    .with_context(|| format!("request_data_upload: path={path}"))
    .err();

    if let Some(err) = maybe_err {
-        tracing::error!(%id, error = ?err, "failed to upload request data");
+        tracing::error!(%id, %path, error = ?err, "failed to upload request data");
    }

    Ok(buffer.writer())
--- a/proxy/src/usage_metrics.rs
+++ b/proxy/src/usage_metrics.rs
@@ -396,13 +396,13 @@ async fn upload_backup_events(
        TimeoutOrCancel::caused_by_cancel,
        FAILED_UPLOAD_WARN_THRESHOLD,
        FAILED_UPLOAD_MAX_RETRIES,
-        "request_data_upload",
+        "usage_metrics_upload",
        cancel,
    )
    .await
    .ok_or_else(|| anyhow::Error::new(TimeoutOrCancel::Cancel))
    .and_then(|x| x)
-    .context("request_data_upload")?;
+    .with_context(|| format!("usage_metrics_upload: path={remote_path}"))?;
    Ok(())
 }

--- a/pyproject.toml
+++ b/pyproject.toml
@@ -43,7 +43,7 @@ websockets = "^12.0"
 clickhouse-connect = "^0.7.16"
 kafka-python = "^2.0.2"
 jwcrypto = "^1.5.6"
-h2 = "^4.1.0"
+h2 = {git = "https://github.com/python-hyper/h2"}
 types-jwcrypto = "^1.5.0.20240925"
 pyyaml = "^6.0.2"
 types-pyyaml = "^6.0.12.20240917"
@@ -94,6 +94,7 @@ target-version = "py311"
 extend-exclude = [
    "vendor/",
    "target/",
+    "test_runner/stubs/", # Autogenerated by mypy's stubgen
 ]
 line-length = 100 # this setting is rather guidance, it won't fail if it can't make the shorter

--- a/safekeeper/client/src/mgmt_api.rs
+++ b/safekeeper/client/src/mgmt_api.rs
@@ -4,7 +4,7 @@
 //! united.

 use reqwest::{IntoUrl, Method, StatusCode};
-use safekeeper_api::models::TimelineStatus;
+use safekeeper_api::models::{TimelineCreateRequest, TimelineStatus};
 use std::error::Error as _;
 use utils::{
    http::error::HttpErrorBody,
@@ -76,6 +76,28 @@ impl Client {
        }
    }

+    pub async fn create_timeline(&self, req: &TimelineCreateRequest) -> Result<TimelineStatus> {
+        let uri = format!(
+            "{}/v1/tenant/{}/timeline/{}",
+            self.mgmt_api_endpoint, req.tenant_id, req.timeline_id
+        );
+        let resp = self.post(&uri, req).await?;
+        resp.json().await.map_err(Error::ReceiveBody)
+    }
+
+    pub async fn delete_timeline(
+        &self,
+        tenant_id: TenantId,
+        timeline_id: TimelineId,
+    ) -> Result<TimelineStatus> {
+        let uri = format!(
+            "{}/v1/tenant/{}/timeline/{}",
+            self.mgmt_api_endpoint, tenant_id, timeline_id
+        );
+        let resp = self.request(Method::DELETE, &uri, ()).await?;
+        resp.json().await.map_err(Error::ReceiveBody)
+    }
+
    pub async fn timeline_status(
        &self,
        tenant_id: TenantId,
@@ -107,6 +129,14 @@ impl Client {
        self.get(&uri).await
    }

+    async fn post<B: serde::Serialize, U: IntoUrl>(
+        &self,
+        uri: U,
+        body: B,
+    ) -> Result<reqwest::Response> {
+        self.request(Method::POST, uri, body).await
+    }
+
    async fn get<U: IntoUrl>(&self, uri: U) -> Result<reqwest::Response> {
        self.request(Method::GET, uri, ()).await
    }
--- a/scripts/comment-test-report.js
+++ b/scripts/comment-test-report.js
@@ -84,6 +84,12 @@ const parseReportJson = async ({ reportJsonUrl, fetch }) => {
                } else {
                    arch = "unknown"
                }
+                let lfcState = ""
+                if (test.parameters.includes("'with-lfc'")) {
+                    lfcState = "with-lfc"
+                } else {
+                    lfcState = "without-lfc"
+                }

                // Removing build type and PostgreSQL version from the test name to make it shorter
                const testName = test.name.replace(new RegExp(`${buildType}-pg${pgVersion}-?`), "").replace("[]", "")
@@ -91,6 +97,7 @@ const parseReportJson = async ({ reportJsonUrl, fetch }) => {
                test.pgVersion = pgVersion
                test.buildType = buildType
                test.arch = arch
+                test.lfcState = lfcState

                if (test.status === "passed") {
                    passedTests[pgVersion][testName].push(test)
@@ -157,7 +164,7 @@ const reportSummary = async (params) => {
                const links = []
                for (const test of tests) {
                    const allureLink = `${reportUrl}#suites/${test.parentUid}/${test.uid}`
-                    links.push(`[${test.buildType}-${test.arch}](${allureLink})`)
+                    links.push(`[${test.buildType}-${test.arch}-${test.lfcState}](${allureLink})`)
                }
                summary += `- \`${testName}\`: ${links.join(", ")}\n`
            }
@@ -188,7 +195,7 @@ const reportSummary = async (params) => {
                    const links = []
                    for (const test of tests) {
                        const allureLink = `${reportUrl}#suites/${test.parentUid}/${test.uid}/retries`
-                        links.push(`[${test.buildType}-${test.arch}](${allureLink})`)
+                        links.push(`[${test.buildType}-${test.arch}-${test.lfcState}](${allureLink})`)
                    }
                    summary += `- \`${testName}\`: ${links.join(", ")}\n`
                }
--- a/scripts/ingest_regress_test_result-new-format.py
+++ b/scripts/ingest_regress_test_result-new-format.py
@@ -134,7 +134,7 @@ def ingest_test_result(
            if p["name"].startswith("__")
        }
        arch = parameters.get("arch", "UNKNOWN").strip("'")
-        lfc = parameters.get("lfc", "False") == "True"
+        lfc = parameters.get("lfc", "without-lfc").strip("'") == "with-lfc"

        build_type, pg_version, unparametrized_name = parse_test_name(test["name"])
        labels = {label["name"]: label["value"] for label in test["labels"]}
--- a/storage_broker/src/bin/storage_broker.rs
+++ b/storage_broker/src/bin/storage_broker.rs
@@ -32,7 +32,6 @@ use tokio::sync::broadcast::error::RecvError;
 use tokio::time;
 use tonic::body::{self, empty_body, BoxBody};
 use tonic::codegen::Service;
-use tonic::transport::server::Connected;
 use tonic::Code;
 use tonic::{Request, Response, Status};
 use tracing::*;
@@ -459,9 +458,10 @@ impl BrokerService for Broker {
        &self,
        request: Request<tonic::Streaming<SafekeeperTimelineInfo>>,
    ) -> Result<Response<()>, Status> {
-        let remote_addr = request
-            .remote_addr()
-            .expect("TCPConnectInfo inserted by handler");
+        let &RemoteAddr(remote_addr) = request
+            .extensions()
+            .get()
+            .expect("RemoteAddr inserted by handler");
        let mut publisher = self.registry.register_publisher(remote_addr);

        let mut stream = request.into_inner();
@@ -484,9 +484,10 @@ impl BrokerService for Broker {
        &self,
        request: Request<SubscribeSafekeeperInfoRequest>,
    ) -> Result<Response<Self::SubscribeSafekeeperInfoStream>, Status> {
-        let remote_addr = request
-            .remote_addr()
-            .expect("TCPConnectInfo inserted by handler");
+        let &RemoteAddr(remote_addr) = request
+            .extensions()
+            .get()
+            .expect("RemoteAddr inserted by handler");
        let proto_key = request
            .into_inner()
            .subscription_key
@@ -537,9 +538,10 @@ impl BrokerService for Broker {
        &self,
        request: Request<SubscribeByFilterRequest>,
    ) -> std::result::Result<Response<Self::SubscribeByFilterStream>, Status> {
-        let remote_addr = request
-            .remote_addr()
-            .expect("TCPConnectInfo inserted by handler");
+        let &RemoteAddr(remote_addr) = request
+            .extensions()
+            .get()
+            .expect("RemoteAddr inserted by handler");
        let proto_filter = request.into_inner();
        let ttid_filter = proto_filter.tenant_timeline_id.as_ref();

@@ -628,6 +630,9 @@ async fn http1_handler(
    Ok(resp)
 }

+#[derive(Clone, Copy)]
+struct RemoteAddr(SocketAddr);
+
 #[tokio::main]
 async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let args = Args::parse();
@@ -687,13 +692,13 @@ async fn main() -> Result<(), Box<dyn std::error::Error>> {
            .max_concurrent_streams(None);

        let storage_broker_server_cloned = storage_broker_server.clone();
-        let connect_info = stream.connect_info();
+        let remote_addr = RemoteAddr(addr);
        let service_fn_ = async move {
            service_fn(move |mut req| {
                // That's what tonic's MakeSvc.call does to pass conninfo to
                // the request handler (and where its request.remote_addr()
                // expects it to find).
-                req.extensions_mut().insert(connect_info.clone());
+                req.extensions_mut().insert(remote_addr);

                // Technically this second clone is not needed, but consume
                // by async block is apparently unavoidable. BTW, error
--- a/storage_controller/Cargo.toml
+++ b/storage_controller/Cargo.toml
@@ -45,12 +45,11 @@ strum_macros.workspace = true

 diesel = { version = "2.2.6", features = [
    "serde_json",
-    "postgres",
-    "r2d2",
    "chrono",
 ] }
+diesel-async = { version = "0.5.2", features = ["postgres", "bb8", "async-connection-wrapper"] }
 diesel_migrations = { version = "2.2.0" }
-r2d2 = { version = "0.8.10" }
+scoped-futures = "0.1.4"

 utils = { path = "../libs/utils/" }
 metrics = { path = "../libs/metrics/" }
--- a/storage_controller/src/main.rs
+++ b/storage_controller/src/main.rs
@@ -308,7 +308,7 @@ async fn async_main() -> anyhow::Result<()> {
    // Validate that we can connect to the database
    Persistence::await_connection(&secrets.database_url, args.db_connect_timeout.into()).await?;

-    let persistence = Arc::new(Persistence::new(secrets.database_url));
+    let persistence = Arc::new(Persistence::new(secrets.database_url).await);

    let service = Service::spawn(config, persistence.clone()).await?;

--- a/storage_controller/src/pageserver_client.rs
+++ b/storage_controller/src/pageserver_client.rs
@@ -2,8 +2,9 @@ use pageserver_api::{
    models::{
        detach_ancestor::AncestorDetached, LocationConfig, LocationConfigListResponse,
        PageserverUtilization, SecondaryProgress, TenantScanRemoteStorageResponse,
-        TenantShardSplitRequest, TenantShardSplitResponse, TimelineArchivalConfigRequest,
-        TimelineCreateRequest, TimelineInfo, TopTenantShardsRequest, TopTenantShardsResponse,
+        TenantShardSplitRequest, TenantShardSplitResponse, TenantWaitLsnRequest,
+        TimelineArchivalConfigRequest, TimelineCreateRequest, TimelineInfo, TopTenantShardsRequest,
+        TopTenantShardsResponse,
    },
    shard::TenantShardId,
 };
@@ -299,4 +300,17 @@ impl PageserverClient {
            self.inner.top_tenant_shards(request).await
        )
    }
+
+    pub(crate) async fn wait_lsn(
+        &self,
+        tenant_shard_id: TenantShardId,
+        request: TenantWaitLsnRequest,
+    ) -> Result<StatusCode> {
+        measured_request!(
+            "wait_lsn",
+            crate::metrics::Method::Post,
+            &self.node_id_label,
+            self.inner.wait_lsn(tenant_shard_id, request).await
+        )
+    }
 }
--- a/storage_controller/src/persistence.rs
+++ b/storage_controller/src/persistence.rs
--- a/storage_controller/src/reconciler.rs
+++ b/storage_controller/src/reconciler.rs
@@ -3,7 +3,7 @@ use crate::persistence::Persistence;
 use crate::{compute_hook, service};
 use pageserver_api::controller_api::{AvailabilityZone, PlacementPolicy};
 use pageserver_api::models::{
-    LocationConfig, LocationConfigMode, LocationConfigSecondary, TenantConfig,
+    LocationConfig, LocationConfigMode, LocationConfigSecondary, TenantConfig, TenantWaitLsnRequest,
 };
 use pageserver_api::shard::{ShardIdentity, TenantShardId};
 use pageserver_client::mgmt_api;
@@ -348,6 +348,32 @@ impl Reconciler {
        Ok(())
    }

+    async fn wait_lsn(
+        &self,
+        node: &Node,
+        tenant_shard_id: TenantShardId,
+        timelines: HashMap<TimelineId, Lsn>,
+    ) -> Result<StatusCode, ReconcileError> {
+        const TIMEOUT: Duration = Duration::from_secs(10);
+
+        let client = PageserverClient::new(
+            node.get_id(),
+            node.base_url(),
+            self.service_config.jwt_token.as_deref(),
+        );
+
+        client
+            .wait_lsn(
+                tenant_shard_id,
+                TenantWaitLsnRequest {
+                    timelines,
+                    timeout: TIMEOUT,
+                },
+            )
+            .await
+            .map_err(|e| e.into())
+    }
+
    async fn get_lsns(
        &self,
        tenant_shard_id: TenantShardId,
@@ -461,6 +487,39 @@ impl Reconciler {
        node: &Node,
        baseline: HashMap<TimelineId, Lsn>,
    ) -> anyhow::Result<()> {
+        // Signal to the pageserver that it should ingest up to the baseline LSNs.
+        loop {
+            match self.wait_lsn(node, tenant_shard_id, baseline.clone()).await {
+                Ok(StatusCode::OK) => {
+                    // Everything is caught up
+                    return Ok(());
+                }
+                Ok(StatusCode::ACCEPTED) => {
+                    // Some timelines are not caught up yet.
+                    // They'll be polled below.
+                    break;
+                }
+                Ok(StatusCode::NOT_FOUND) => {
+                    // None of the timelines are present on the pageserver.
+                    // This is correct if they've all been deleted, but
+                    // let let the polling loop below cross check.
+                    break;
+                }
+                Ok(status_code) => {
+                    tracing::warn!(
+                        "Unexpected status code ({status_code}) returned by wait_lsn endpoint"
+                    );
+                    break;
+                }
+                Err(e) => {
+                    tracing::info!("🕑 Can't trigger LSN wait on {node} yet, waiting ({e})",);
+                    tokio::time::sleep(Duration::from_millis(500)).await;
+                    continue;
+                }
+            }
+        }
+
+        // Poll the LSNs until they catch up
        loop {
            let latest = match self.get_lsns(tenant_shard_id, node).await {
                Ok(l) => l,
--- a/storage_scrubber/src/pageserver_physical_gc.rs
+++ b/storage_scrubber/src/pageserver_physical_gc.rs
@@ -8,6 +8,8 @@ use crate::checks::{
 };
 use crate::metadata_stream::{stream_tenant_timelines, stream_tenants};
 use crate::{init_remote, BucketConfig, NodeKind, RootTarget, TenantShardTimelineId, MAX_RETRIES};
+use async_stream::try_stream;
+use futures::future::Either;
 use futures_util::{StreamExt, TryStreamExt};
 use pageserver::tenant::remote_timeline_client::index::LayerFileMetadata;
 use pageserver::tenant::remote_timeline_client::manifest::OffloadedTimelineManifest;
@@ -578,7 +580,7 @@ async fn gc_timeline(
    target: &RootTarget,
    mode: GcMode,
    ttid: TenantShardTimelineId,
-    accumulator: &Arc<std::sync::Mutex<TenantRefAccumulator>>,
+    accumulator: &std::sync::Mutex<TenantRefAccumulator>,
    tenant_manifest_info: Arc<Option<RemoteTenantManifestInfo>>,
 ) -> anyhow::Result<GcSummary> {
    let mut summary = GcSummary::default();
@@ -721,9 +723,9 @@ pub async fn pageserver_physical_gc(

    let remote_client = Arc::new(remote_client);
    let tenants = if tenant_shard_ids.is_empty() {
-        futures::future::Either::Left(stream_tenants(&remote_client, &target))
+        Either::Left(stream_tenants(&remote_client, &target))
    } else {
-        futures::future::Either::Right(futures::stream::iter(tenant_shard_ids.into_iter().map(Ok)))
+        Either::Right(futures::stream::iter(tenant_shard_ids.into_iter().map(Ok)))
    };

    // How many tenants to process in parallel.  We need to be mindful of pageservers
@@ -731,16 +733,16 @@ pub async fn pageserver_physical_gc(
    const CONCURRENCY: usize = 32;

    // Accumulate information about each tenant for cross-shard GC step we'll do at the end
-    let accumulator = Arc::new(std::sync::Mutex::new(TenantRefAccumulator::default()));
+    let accumulator = std::sync::Mutex::new(TenantRefAccumulator::default());
+
+    // Accumulate information about how many manifests we have GCd
+    let manifest_gc_summary = std::sync::Mutex::new(GcSummary::default());

    // Generate a stream of TenantTimelineId
-    enum GcSummaryOrContent<T> {
-        Content(T),
-        GcSummary(GcSummary),
-    }
    let timelines = tenants.map_ok(|tenant_shard_id| {
        let target_ref = &target;
        let remote_client_ref = &remote_client;
+        let manifest_gc_summary_ref = &manifest_gc_summary;
        async move {
            let gc_manifest_result = gc_tenant_manifests(
                remote_client_ref,
@@ -757,55 +759,48 @@ pub async fn pageserver_physical_gc(
                    (GcSummary::default(), None)
                }
            };
+            manifest_gc_summary_ref
+                .lock()
+                .unwrap()
+                .merge(summary_from_manifest);
            let tenant_manifest_arc = Arc::new(tenant_manifest_opt);
-            let summary_from_manifest = Ok(GcSummaryOrContent::<(_, _)>::GcSummary(
-                summary_from_manifest,
-            ));
-            stream_tenant_timelines(remote_client_ref, target_ref, tenant_shard_id)
-                .await
-                .map(|stream| {
-                    stream
-                        .zip(futures::stream::iter(std::iter::repeat(
-                            tenant_manifest_arc,
-                        )))
-                        .map(|(ttid_res, tenant_manifest_arc)| {
-                            ttid_res.map(move |ttid| {
-                                GcSummaryOrContent::Content((ttid, tenant_manifest_arc))
-                            })
-                        })
-                        .chain(futures::stream::iter([summary_from_manifest].into_iter()))
-                })
+            let mut timelines = Box::pin(
+                stream_tenant_timelines(remote_client_ref, target_ref, tenant_shard_id).await?,
+            );
+            Ok(try_stream! {
+                while let Some(ttid_res) = timelines.next().await {
+                    let ttid = ttid_res?;
+                    yield (ttid, tenant_manifest_arc.clone());
+                }
+            })
        }
    });
-    let timelines = std::pin::pin!(timelines.try_buffered(CONCURRENCY));
-    let timelines = timelines.try_flatten();

    let mut summary = GcSummary::default();
-
-    // Drain futures for per-shard GC, populating accumulator as a side effect
    {
-        let timelines = timelines.map_ok(|summary_or_ttid| match summary_or_ttid {
-            GcSummaryOrContent::Content((ttid, tenant_manifest_arc)) => {
-                futures::future::Either::Left(gc_timeline(
-                    &remote_client,
-                    &min_age,
-                    &target,
-                    mode,
-                    ttid,
-                    &accumulator,
-                    tenant_manifest_arc,
-                ))
-            }
-            GcSummaryOrContent::GcSummary(gc_summary) => {
-                futures::future::Either::Right(futures::future::ok(gc_summary))
-            }
+        let timelines = std::pin::pin!(timelines.try_buffered(CONCURRENCY));
+        let timelines = timelines.try_flatten();
+
+        let timelines = timelines.map_ok(|(ttid, tenant_manifest_arc)| {
+            gc_timeline(
+                &remote_client,
+                &min_age,
+                &target,
+                mode,
+                ttid,
+                &accumulator,
+                tenant_manifest_arc,
+            )
        });
        let mut timelines = std::pin::pin!(timelines.try_buffered(CONCURRENCY));

+        // Drain futures for per-shard GC, populating accumulator as a side effect
        while let Some(i) = timelines.next().await {
            summary.merge(i?);
        }
    }
+    // Streams are lazily evaluated, so only now do we have access to the inner object
+    summary.merge(manifest_gc_summary.into_inner().unwrap());

    // Execute cross-shard GC, using the accumulator's full view of all the shards built in the per-shard GC
    let Some(client) = controller_client else {
@@ -813,8 +808,7 @@ pub async fn pageserver_physical_gc(
        return Ok(summary);
    };

-    let (ancestor_shards, ancestor_refs) = Arc::into_inner(accumulator)
-        .unwrap()
+    let (ancestor_shards, ancestor_refs) = accumulator
        .into_inner()
        .unwrap()
        .into_gc_ancestors(client, &mut summary)
--- a/test_runner/conftest.py
+++ b/test_runner/conftest.py
@@ -15,4 +15,5 @@ pytest_plugins = (
    "fixtures.compare_fixtures",
    "fixtures.slow",
    "fixtures.reruns",
+    "fixtures.fast_import",
 )
--- a/test_runner/fixtures/common_types.py
+++ b/test_runner/fixtures/common_types.py
@@ -208,6 +208,10 @@ class ShardIndex:
            shard_count=int(input[2:4], 16),
        )

+    @property
+    def is_sharded(self) -> bool:
+        return self.shard_count != 0
+

 class TenantShardId:
    def __init__(self, tenant_id: TenantId, shard_number: int, shard_count: int):
--- a/test_runner/fixtures/fast_import.py
+++ b/test_runner/fixtures/fast_import.py
@@ -0,0 +1,104 @@
+import os
+import shutil
+import subprocess
+import tempfile
+from collections.abc import Iterator
+from pathlib import Path
+
+import pytest
+
+from fixtures.log_helper import log
+from fixtures.neon_cli import AbstractNeonCli
+from fixtures.pg_version import PgVersion
+
+
+class FastImport(AbstractNeonCli):
+    COMMAND = "fast_import"
+    cmd: subprocess.CompletedProcess[str] | None = None
+
+    def __init__(
+        self,
+        extra_env: dict[str, str] | None,
+        binpath: Path,
+        pg_distrib_dir: Path,
+        pg_version: PgVersion,
+        workdir: Path,
+    ):
+        if extra_env is None:
+            env_vars = {}
+        else:
+            env_vars = extra_env.copy()
+
+        if not (binpath / self.COMMAND).exists():
+            raise Exception(f"{self.COMMAND} binary not found at '{binpath}'")
+        super().__init__(env_vars, binpath)
+
+        pg_dir = pg_distrib_dir / pg_version.v_prefixed
+        self.pg_distrib_dir = pg_distrib_dir
+        self.pg_version = pg_version
+        self.pg_bin = pg_dir / "bin"
+        if not (self.pg_bin / "postgres").exists():
+            raise Exception(f"postgres binary was not found at '{self.pg_bin}'")
+        self.pg_lib = pg_dir / "lib"
+        if env_vars.get("LD_LIBRARY_PATH") is not None:
+            self.pg_lib = Path(env_vars["LD_LIBRARY_PATH"])
+        elif os.getenv("LD_LIBRARY_PATH") is not None:
+            self.pg_lib = Path(str(os.getenv("LD_LIBRARY_PATH")))
+        if not workdir.exists():
+            raise Exception(f"Working directory '{workdir}' does not exist")
+        self.workdir = workdir
+
+    def run(
+        self,
+        pg_port: int,
+        source_connection_string: str | None = None,
+        s3prefix: str | None = None,
+        interactive: bool = False,
+    ) -> subprocess.CompletedProcess[str]:
+        if self.cmd is not None:
+            raise Exception("Command already executed")
+        args = [
+            f"--pg-bin-dir={self.pg_bin}",
+            f"--pg-lib-dir={self.pg_lib}",
+            f"--pg-port={pg_port}",
+            f"--working-directory={self.workdir}",
+        ]
+        if source_connection_string is not None:
+            args.append(f"--source-connection-string={source_connection_string}")
+        if s3prefix is not None:
+            args.append(f"--s3-prefix={s3prefix}")
+        if interactive:
+            args.append("--interactive")
+
+        self.cmd = self.raw_cli(args)
+        return self.cmd
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, *args):
+        if self.workdir.exists():
+            shutil.rmtree(self.workdir)
+
+
+@pytest.fixture(scope="function")
+def fast_import(
+    pg_version: PgVersion,
+    test_output_dir: Path,
+    neon_binpath: Path,
+    pg_distrib_dir: Path,
+) -> Iterator[FastImport]:
+    workdir = Path(tempfile.mkdtemp())
+    with FastImport(None, neon_binpath, pg_distrib_dir, pg_version, workdir) as fi:
+        yield fi
+
+        if fi.cmd is None:
+            return
+
+        # dump stdout & stderr into test log dir
+        with open(test_output_dir / "fast_import.stdout", "w") as f:
+            f.write(fi.cmd.stdout)
+        with open(test_output_dir / "fast_import.stderr", "w") as f:
+            f.write(fi.cmd.stderr)
+
+        log.info("Written logs to %s", test_output_dir)
--- a/test_runner/fixtures/metrics.py
+++ b/test_runner/fixtures/metrics.py
@@ -126,12 +126,8 @@ PAGESERVER_GLOBAL_METRICS: tuple[str, ...] = (
    "pageserver_page_cache_read_accesses_total",
    "pageserver_page_cache_size_current_bytes",
    "pageserver_page_cache_size_max_bytes",
-    "pageserver_getpage_reconstruct_seconds_bucket",
-    "pageserver_getpage_reconstruct_seconds_count",
-    "pageserver_getpage_reconstruct_seconds_sum",
    *[f"pageserver_basebackup_query_seconds_{x}" for x in ["bucket", "count", "sum"]],
    *histogram("pageserver_smgr_query_seconds_global"),
-    *histogram("pageserver_getpage_get_reconstruct_data_seconds"),
    *histogram("pageserver_wait_lsn_seconds"),
    *histogram("pageserver_remote_operation_seconds"),
    *histogram("pageserver_io_operations_seconds"),
--- a/test_runner/fixtures/neon_cli.py
+++ b/test_runner/fixtures/neon_cli.py
@@ -523,6 +523,7 @@ class NeonLocalCli(AbstractNeonCli):
        remote_ext_config: str | None = None,
        pageserver_id: int | None = None,
        allow_multiple: bool = False,
+        create_test_user: bool = False,
        basebackup_request_tries: int | None = None,
        env: dict[str, str] | None = None,
    ) -> subprocess.CompletedProcess[str]:
@@ -544,6 +545,8 @@ class NeonLocalCli(AbstractNeonCli):
            args.extend(["--pageserver-id", str(pageserver_id)])
        if allow_multiple:
            args.extend(["--allow-multiple"])
+        if create_test_user:
+            args.extend(["--create-test-user"])

        res = self.raw_cli(args, extra_env_vars)
        res.check_returncode()
--- a/Show More
+++ b/Show More