release-19.1: storage: rate-limit AddSST requests #36444
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
We've been seeing extremely high latency for foreground traffic during bulk index backfills, because AddSST requests into non-empty ranges can be expensive, and write requests that are queued behind an AddSST request for an overlapping span can get stuck waiting for multiple seconds. This PR limits the number of concurrent AddSST requests for a single store, determined by a new cluster setting, `kv.bulk_io_write.concurrent_addsstable_requests`, to decrease the impact of index backfills on foreground writes. (It also decreases the risk of writing too many L0 files to RocksDB at once, which causes stalls.) Release note (general change): Add a new cluster setting, `kv.bulk_io_write.concurrent_addsstable_requests`, which limits the number of SSTables that can be added concurrently during bulk operations.
|
This change is |
dt
approved these changes
Apr 2, 2019
|
bors r+ |
|
bors r- |
Canceled |
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Outdated suggestions cannot be applied.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.
Backport 1/1 commits from #36403.
/cc @cockroachdb/release
We've been seeing extremely high latency for foreground traffic during bulk
index backfills, because AddSST requests into non-empty ranges can be
expensive, and write requests that are queued behind an AddSST request for an
overlapping span can get stuck waiting for multiple seconds. This PR limits the
number of concurrent AddSST requests for a single store, determined by a new
cluster setting,
kv.bulk_io_write.concurrent_addsstable_requests, to decreasethe impact of index backfills on foreground writes. (It also decreases the risk
of writing too many L0 files to RocksDB at once, which causes stalls.)
Fixes #36430
Release note (general change): Add a new cluster setting,
kv.bulk_io_write.concurrent_addsstable_requests, which limits the number ofSSTables that can be added concurrently during bulk operations.