feat: Coalesce small batches before shuffle write #3234
Closed
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
…ency This change adds batch coalescing before shuffle writes to reduce per-batch overhead and improve vectorization efficiency. When enabled, small columnar batches are combined until they reach the target batch size before being processed by the shuffle writer. Benefits observed in TPC-H Q18 benchmarks: - 10.9% overall query time improvement - Significantly reduced GC pressure (Stage 26: 3,602ms -> 56ms GC time) - Better vectorization efficiency for downstream operators New configuration options: - spark.comet.shuffle.resizeBatches.input: Coalesce batches before shuffle write (default: false) - spark.comet.shuffle.resizeBatches.output: Coalesce batches after shuffle read (default: true) The native planner now wraps shuffle input with DataFusion's CoalesceBatchesExec when spark.comet.shuffle.resizeBatches.input is enabled. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
andygrove
force-pushed
the
coalesce-batches-shuffle
branch
from
fdc9074 to
5cccc1b
Compare
andygrove
changed the title
andygrove
changed the title
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Outdated suggestions cannot be applied.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.
Summary
This PR adds batch coalescing before shuffle writes to reduce per-batch overhead and improve vectorization efficiency. When enabled, small columnar batches are combined until they reach the target batch size before being processed by the shuffle writer.
Key changes:
spark.comet.shuffle.resizeBatches.inputconfig to enable coalescing batches before shuffle writespark.comet.shuffle.resizeBatches.outputconfig for coalescing after shuffle readCoalesceBatchesExecwhen input coalescing is enabledCometBatchCoalescerScala class for output-side batch coalescingTest plan
spark.comet.shuffle.resizeBatches.input=true🤖 Generated with Claude Code