John Khvatov 33cb9a68f7 pageserver: Reduce tracing overhead in timeline::get (#6115) ...

## Problem

Compaction process (specifically the image layer reconstructions part)
is lagging behind wal ingest (at speed ~10-15MB/s) for medium-sized
tenants (30-50GB). CPU profile shows that significant amount of time
(see flamegraph) is being spent in `tracing::span::Span::new`.

mainline (commit: 0ba4cae491):

![reconstruct-mainline-0ba4cae491c2](https://github.com/neondatabase/neon/assets/289788/ebfd262e-5c97-4858-80c7-664a1dbcc59d)

## Summary of changes

By lowering the tracing level in get_value_reconstruct_data and
get_or_maybe_download from info to debug, we can reduce the overhead of
span creation in prod environments. On my system, this sped up the image
reconstruction process by 60% (from 14500 to 23160 page reconstruction
per sec)

pr:

![reconstruct-opt-2](https://github.com/neondatabase/neon/assets/289788/563a159b-8f2f-4300-b0a1-6cd66e7df769)


`create_image_layers()` (it's 1 CPU bound here) mainline vs pr:

![image](https://github.com/neondatabase/neon/assets/289788/a981e3cb-6df9-4882-8a94-95e99c35aa83)

2023-12-18 13:33:23 +00:00

..

benches

pageserver: refactor TenantId to TenantShardId in Tenant & Timeline (#5957)

2023-11-29 14:52:35 +00:00

client

pageserver: add method to dump keyspace in mgmt api client (#6145)

2023-12-16 10:52:48 +00:00

ctl

pagectl: add subcommand to rewrite layer file summary (#5933)

2023-11-30 11:34:30 +00:00

src

pageserver: Reduce tracing overhead in timeline::get (#6115)

2023-12-18 13:33:23 +00:00

test_data

Add walingest test (#5892)

2023-11-28 12:50:53 -05:00

Cargo.toml

pageserver: heatmap uploads (#6050)

2023-12-14 13:09:24 +00:00