Repository navigation
Conversation
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Reads admitted in separate polls currently start separate quorum rounds even while an earlier round is still waiting for replies. Keep one unconfirmed round per group and collect later arrivals in the existing bounded, unsent batch. Issue that batch's fresh read-index barrier in the same poll that confirms the preceding round.
Later arrivals cannot use the preceding round's quorum. Request capacity, apply barriers, term checks and durable write acknowledgements remain unchanged. Leadership loss retires both the sent round and its queued successor.
The native regression admits 17 reads in separate polls while delaying the first quorum reply. Before this change it emits 17 read contexts; afterward it emits one, followed immediately by one fresh context. The first reply completes only the first read. All reads and another append finish, and all three original engines reopen with both full payloads.
Draft: a mixed ingestion, update and point-read run passed its full-value and bounded-progress checks, but did not show a meaningful throughput improvement; point-read tail latency and missed read arrivals worsened. Keep this PR in draft pending a better throughput/latency result. Coalescing overlapping rounds does not remove the need for fresh quorums on sequential reads or establish an end-to-end throughput gain.
Validation
Refs #378.