kgo: unlinger partitions in ProduceSync to avoid linger delay #1236
+157
−23
Conversation
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
Add -b/--bench and -c/--count flags for running Go benchmarks. Derive kfake and kgo directories from the script location rather than using hardcoded paths. Co-Authored-By: Claude Opus 4.6 <[email protected]>
ProduceSync now stops lingering and triggers an immediate drain on all partitions that records were produced to. This avoids the caller waiting unnecessarily for linger timers (default 10ms since v1.20.0) when synchronously waiting for results. After each Produce call, the record's Partition field is already set (see bufferRecord), so we collect unique recBufs inline and unlinger them after all records are enqueued. A [16]*recBuf stack array with linear dedup avoids heap allocation in the common case. Closes #1195 Co-Authored-By: Claude Opus 4.6 <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
ProduceSync now stops lingering and triggers an immediate drain on
all partitions that records were produced to. This avoids the caller
waiting unnecessarily for linger timers (default 10ms since v1.20.0)
when synchronously waiting for results.
After each Produce call, the record's Partition field is already set
(see bufferRecord), so we collect unique recBufs inline and unlinger
them after all records are enqueued. A [16]*recBuf stack array with
linear dedup avoids heap allocation in the common case.
Closes #1195