pageserver: add per-timeline read amp histogram #10566
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
erikgrinaker
changed the title
skyzh
approved these changes
Jan 29, 2025
7414 tests run: 7063 passed, 0 failed, 351 skipped (full report)Flaky tests (8)Postgres 17
Code coverage* (full report)
* collected from Rust tests only The comment gets automatically updated with the latest test results
8617a8d at 2025-01-30T11:32:48.632Z :recycle: |
erikgrinaker
force-pushed
the
erik/layers-per-read-metric
branch
from
d302445 to
f368c55
Compare
erikgrinaker
changed the base branch from
main
to
erik/layers-per-read-global
|
I split out the fix of the global |
erikgrinaker
changed the title
erikgrinaker
changed the title
github-merge-queue bot
pushed a commit
that referenced
this pull request
Jan 30, 2025
## Problem The current global `pageserver_layers_visited_per_vectored_read_global` metric does not appear to accurately measure read amplification. It divides the layer count by the number of reads in a batch, but this means that e.g. 10 reads with 100 L0 layers will only measure a read amp of 10 per read, while the actual read amp was 100. While the cost of layer visits are amortized across the batch, and some layers may not intersect with a given key, each visited layer contributes directly to the observed latency for every read in the batch, which is what we care about. Touches neondatabase/cloud#23283. Extracted from #10566. ## Summary of changes * Count the number of layers visited towards each read in the batch, instead of the average across the batch. * Rename `pageserver_layers_visited_per_vectored_read_global` to `pageserver_layers_per_read_global`. * Reduce the read amp log warning threshold down from 512 to 100.
erikgrinaker
force-pushed
the
erik/layers-per-read-metric
branch
from
f368c55 to
8617a8d
Compare
|
To reduce the cardinality, I lowered the resolution here to: The global metric has slightly better resolution: I think that should be enough to give us a rough idea of timeline read amp. If the global metric shows e.g. >256, we can likely narrow it down to a handful of tenants with the per-timeline histogram. I'm going to merge this as-is, we can always change/remove it later if we need to. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Problem
We don't have per-timeline observability for read amplification.
Touches https://github.com/neondatabase/cloud/issues/23283.
Summary of changes
Add a per-timeline
pageserver_layers_per_readhistogram.NB: per-timeline histograms are expensive, but probably worth it in this case.