Node-level shard lookup for persist requests #6118
Merged
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
guilload
force-pushed
the
nadav/node-specific-lookup
branch
2 times, most recently
from
894825b to
53b64ff
Compare
guilload
force-pushed
the
nadav/node-specific-lookup
branch
from
53b64ff to
34c664b
Compare
guilload
approved these changes
Jan 27, 2026
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Outdated suggestions cannot be applied.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.
Once requests are routed to nodes, we can have the node try to find the shard with the most capacity, to better load balance requests. Since the node is the source of truth for tracking its own shards' capacity, this should be a minor to moderate optimization to better pack persist requests between nodes.
This requires removing the "Tracked" state on the publish tracker and moving that logic into the persisted state. It should be a no-op, but introduces the (seemingly impossible) chance that if a publish arrives before a persist, we'd drop it.
The other "hack" here is that the persist subrequests still target an individual shard, but that shard isn't necessarily the one we end up persisting to. This isn't a problem in the literal sense because PersistSuccesses do report the correct shard, and the publish tracker keeps track per subrequest and then updates from the shard within the subrequest. But there will be some logical misdirection there.
Updated tests to pass. Next step is testing in staging.