Fix Possible Congestion Scenario in SPM #12230
Closed
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
github-actions
bot
added
physical-expr
berkaysynnada
force-pushed
the
feature/spm-cursor-init
branch
from
0449f47 to
4382945
Compare
|
Follow on PR is #12302 |
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Outdated suggestions cannot be applied.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.
Which issue does this PR close?
Closes #.
Rationale for this change
During the initialization of cursors of SPM (
SortPreservingMergeExec), it needs to getPoll::Ready()for each partition, to initiate the next partition cursor. We have experienced a scenario in our downstream where SPM is polling 1st partition continuously but never get aPoll::Ready(). Those polls get all upstream operator buffers and channels grow. However, if SPM have skipped that partition and tried to initiate other partitions before retry, we did not experience this problem (congestion in partition 1 is cleared once the other partitions are pulled).What changes are included in this PR?
SPM does not wait initiating the next cursor partition. Instead, it iterates over all partitions continuously whether it is OK or not, until they are all successfully initiated.
Are these changes tested?
Yes, with a unit test. The test depends on a timeout duration to sense deadlock but I could also add a memory usage test.
Are there any user-facing changes?