Replace the layer array and linear search with R-tree
So far, the in-memory layer map that holds information about layer
files that exist, has used a simple Vec, in no particular order, to
hold information about all the layers. That obviously doesn't scale
very well; with thousands of layer files the linear search was
consuming a lot of CPU. Replace it with a two-dimensional R-tree, with
Key and LSN ranges as the dimensions.
For the R-tree, use the 'rstar' crate. To be able to use that, we
convert the Keys and LSNs into 256-bit integers. 64 bits would be
enough to represent LSNs, and 128 bits would be enough to represent
Keys. However, we use 256 bits, because rstar internally performs
multiplication to calculate the area of rectangles, and the result of
multiplying two 128 bit integers doesn't necessarily fit in 128 bits,
causing integer overflow and, if overflow-checks are enabled,
panic. To avoid that, we use 256 bit integers.
Add a performance test that creates a lot of layer files, to
demonstrate the benefit.
Instead of spawning helper threads, we now use Tokio tasks. There
are multiple Tokio runtimes, for different kinds of tasks. One for
serving libpq client connections, another for background operations
like GC and compaction, and so on. That's not strictly required, we
could use just one runtime, but with this you can still get an
overview of what's happening with "top -H".
There's one subtle behavior in how TenantState is updated. Before this
patch, if you deleted all timelines from a tenant, its GC and
compaction loops were stopped, and the tenant went back to Idle
state. We no longer do that. The empty tenant stays Active. The
changes to test_tenant_tasks.py are related to that.
There's still plenty of synchronous code and blocking. For example, we
still use blocking std::io functions for all file I/O, and the
communication with WAL redo processes is still uses low-level unix
poll(). We might want to rewrite those later, but this will do for
now. The model is that local file I/O is considered to be fast enough
that blocking - and preventing other tasks running in the same thread -
is acceptable.
This depends on a hacked version of the 'pprof-rs' crate. Because of
that, it's under an optional 'profiling' feature. It is disabled by
default, but enabled for release builds in CircleCI config. It doesn't
currently work on macOS.
The flamegraph is written to 'flamegraph.svg' in the pageserver
workdir when the 'pageserver' process exits.
Add a performance test that runs the perf_pgbench test, with profiling
enabled.
We now use a page cache for those, instead of slurping the whole index into
memory.
Fixes https://github.com/zenithdb/zenith/issues/1356
This is a backwards-incompatible change to the storage format, so
bump STORAGE_FORMAT_VERSION.
This introduces two new abstraction layers for I/O:
- Block I/O, and
- Blob I/O.
The BlockReader trait abstracts a file or something else that can be read
in 8kB pages. It is implemented by EphemeralFiles, and by a new
FileBlockReader struct that allows reading arbitrary VirtualFiles in that
manner, utilizing the page cache.
There is also a new BlockCursor struct that works as a cursor over a
BlockReader. When you create a BlockCursor and read the first page using
it, it keeps the reference to the page. If you access the same page again,
it avoids going to page cache and quickly returns the same page again.
That can save a lot of lookups in the page cache if you perform multiple
reads.
The Blob-oriented API allows reading and writing "blobs" of arbitrary
length. It is a layer on top of the block-oriented API. When you write
a blob with the write_blob() function, it writes a length field
followed by the actual data to the underlying block storage, and
returns the offset where the blob was stored. The blob can be
retrieved later using the offset.
Finally, this replaces the I/O code in image-, delta-, and in-memory
layers to use the new abstractions. These replace the 'bookfile'
crate.
This is a backwards-incompatible change to the storage format.
workspace_hack is needed to avoid recompilation when different crates
inside the workspace depend on the same packages but with different
features being enabled. Problem occurs when you build crates separately
one by one. So this is irrelevant to our CI setup because there we build
all binaries at once, but it may be relevant for local development.
this also changes cargo's resolver version to 2
This is a backwards-incompatible change. The new pageserver cannot
read repositories created with an old pageserver binary, or vice
versa.
Simplify Repository to a value-store
------------------------------------
Move the responsibility of tracking relation metadata, like which
relations exist and what are their sizes, from Repository to a new
module, pgdatadir_mapping.rs. The interface to Repository is now a
simple key-value PUT/GET operations.
It's still not any old key-value store though. A Repository is still
responsible from handling branching, and every GET operation comes
with an LSN.
Mapping from Postgres data directory to keys/values
---------------------------------------------------
All the data is now stored in the key-value store. The
'pgdatadir_mapping.rs' module handles mapping from PostgreSQL objects
like relation pages and SLRUs, to key-value pairs.
The key to the Repository key-value store is a Key struct, which
consists of a few integer fields. It's wide enough to store a full
RelFileNode, fork and block number, and to distinguish those from
metadata keys.
'pgdatadir_mapping.rs' is also responsible for maintaining a
"partitioning" of the keyspace. Partitioning means splitting the
keyspace so that each partition holds a roughly equal number of keys.
The partitioning is used when new image layer files are created, so
that each image layer file is roughly the same size.
The partitioning is also responsible for reclaiming space used by
deleted keys. The Repository implementation doesn't have any explicit
support for deleting keys. Instead, the deleted keys are simply
omitted from the partitioning, and when a new image layer is created,
the omitted keys are not copied over to the new image layer. We might
want to implement tombstone keys in the future, to reclaim space
faster, but this will work for now.
Changes to low-level layer file code
------------------------------------
The concept of a "segment" is gone. Each layer file can now store an
arbitrary range of Keys.
Checkpointing, compaction
-------------------------
The background tasks are somewhat different now. Whenever
checkpoint_distance is reached, the WAL receiver thread "freezes" the
current in-memory layer, and creates a new one. This is a quick
operation and doesn't perform any I/O yet. It then launches a
background "layer flushing thread" to write the frozen layer to disk,
as a new L0 delta layer. This mechanism takes care of durability. It
replaces the checkpointing thread.
Compaction is a new background operation that takes a bunch of L0
delta layers, and reshuffles the data in them. It runs in a separate
compaction thread.
Deployment
----------
This also contains changes to the ansible scripts that enable having
multiple different pageservers running at the same time in the staging
environment. We will use that to keep an old version of the pageserver
running, for clusters created with the old version, at the same time
with a new pageserver with the new binary.
Author: Heikki Linnakangas
Author: Konstantin Knizhnik <knizhnik@zenith.tech>
Author: Andrey Taranik <andrey@zenith.tech>
Reviewed-by: Matthias Van De Meent <matthias@zenith.tech>
Reviewed-by: Bojan Serafimov <bojan@zenith.tech>
Reviewed-by: Konstantin Knizhnik <knizhnik@zenith.tech>
Reviewed-by: Anton Shyrabokau <antons@zenith.tech>
Reviewed-by: Dhammika Pathirana <dham@zenith.tech>
Reviewed-by: Kirill Bulatov <kirill@zenith.tech>
Reviewed-by: Anastasia Lubennikova <anastasia@zenith.tech>
Reviewed-by: Alexey Kondratov <alexey@zenith.tech>
to pass current_timeline_size to compute node
Put standby_status_update fields into ZenithFeedback and send them as one message.
Pass values sizes together with keys in ZenithFeedback message.
'anyhow' crate can include a backtrace in all errors, when the
'backtrace' feature is enabled. Enable it, and change the places that used
'{:#}' or '{}' to '{:?}', so that the backtrace is printed.
This patch allows to shutdown wal receiver when there are no messages
and wal receiver is blocked inside tokio-postgres. In this case it
cannot check the shutdown flag.
This patch switches to use async interface of tokio-postgres directly
without sync wrappers. It opens the possibility to use tokio::select!
between the phsycal_stream.next() and a shutdown channel readiness to
interrupt replication process.
Also this allows to shutdown only particular wal receiver without
using global shutdown_requested flag.
We depends on rustls in postgres_backend anyway, so might as well use it
for all TLS stuff. Seems better to depend on only one library both from a
security point of view, and because fewer dependencies means less code to
compile. With this commit, we no longer depend on OpenSSL.
0.28.0 includes two changes I submitted to upstream:
- Add support for older ListObjects API, needed to use rust-s3 with Google
Cloud Storage: https://github.com/durch/rust-s3/pull/229
- If file is smaller than one chunk, don't initiate multi-part upload.
https://github.com/durch/rust-s3/pull/228
These are not critical for Zenith right now, but let's stay up-to-date.
This patch introduces fixes for several problems affecting
LLVM-based code coverage:
* Daemonizing parent processes should call _exit() to prevent
coverage data file corruption (*.profraw) due to concurrent writes.
* Implement proper shutdown handlers in safekeeper.
Currently, whenever a page version is needed from an image or delta
layer, we open the file and read and parse the bookfile headers. That's
pretty expensive. To reduce the overhead, introduce a cache of open file
descriptors, and use that to cache the Book objects so that we don't need
to read the metadata on every access.
The tokio futures added some overhead, so switch to plain non-blocking
I/O with poll(). In a simple pgbench test on my laptop (select-only
queries, scale-factor 1 `pgbench -P1 -T50 -S`), this gives about 10%
improvement, from about 4300 TPS to 4800 TPS.
This calculation is not that heavy but it is needed only in tests, and
in case the number of tenants/timelines is high the calculation can take
noticeable time.
Resolves https://github.com/zenithdb/zenith/issues/804
