Api cache optimizations - #42
Draft
fernandodeluret wants to merge 2 commits into
Draft
fernandodeluret wants to merge 2 commits into
fernandodeluret wants to merge 2 commits into
Conversation
…locked time not do any deallocation, optimize path for small not cached requests
ArthurPaivaT
approved these changes
Aug 27, 2026
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Changes:
GPA cache entries no longer pin ~64 KB streaming TCP chunks. Fresh accounts are copied into tight per-account allocations so resident memory matches accounted cache size. We pay an small copy cost on new accounts going into the cache (for the ones already there no added cost) and get in reward more memory efficient cache size (because we don't keep unnecesarily big 64kb chunks alive).
Cache insert (finalize_query) runs on the blocking pool instead of the request task. The client is no longer waiting on map build, write lock, or eviction.
The cache write lock no longer deallocates. Replaced and evicted entries are dropped after the lock is released.
Queries under min_bytes_per_query are rejected with a cheap size check and never enter the insert path.
Finalize jobs skip a query if a newer slot is already cached, and expose inflight / skipped metrics for that path.