perf: Coalesce broadcast exchange batches before broadcasting#3703
Draft
mbutrovich wants to merge 2 commits intoapache:mainfrom
Draft
perf: Coalesce broadcast exchange batches before broadcasting#3703mbutrovich wants to merge 2 commits intoapache:mainfrom
mbutrovich wants to merge 2 commits intoapache:mainfrom
Conversation
CometBroadcastExchangeExec previously broadcast an Array[ChunkedByteBuffer] with one entry per source partition. Each consumer partition independently deserialized all entries, creating a separate compression codec and Arrow IPC reader per entry. For broadcasts with many source partitions, this produced large per-task overhead in the hash join build-side collection. Decode and concatenate all broadcast batches into a single ChunkedByteBuffer on the driver using VectorSchemaRootAppender before broadcasting. Falls back to per-batch serialization if dictionary-encoded vectors are present.
Contributor
Author
|
In addition to TPC-H, I will see if we can come up with a microbenchmark for this. |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
Closes #3692.
Rationale for this change
CometBroadcastExchangeExec previously broadcast an Array[ChunkedByteBuffer] with one entry per source partition. Each consumer partition independently deserialized all entries, creating a separate compression codec and Arrow IPC reader per entry. For broadcasts with many source partitions, this produced large per-task overhead in the hash join build-side collection.
What changes are included in this PR?
Decode and concatenate all broadcast batches into a single ChunkedByteBuffer on the driver using VectorSchemaRootAppender before broadcasting. Falls back to per-batch serialization if dictionary-encoded vectors are present.
How are these changes tested?
Existing tests.