[SPARK-56319] ShuffleExternalSorter should reuse write buffer#55136
Open
leixm wants to merge 1 commit intoapache:masterfrom
Open
[SPARK-56319] ShuffleExternalSorter should reuse write buffer#55136leixm wants to merge 1 commit intoapache:masterfrom
leixm wants to merge 1 commit intoapache:masterfrom
Conversation
Contributor
Author
|
@cloud-fan Can you help review? |
Contributor
Author
|
cc @LuciferYang Can you help review this PR? |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changes were proposed in this pull request?
[SPARK-56319] ShuffleExternalSorter should reuse write buffer
Why are the changes needed?
In ShuffleExternalSorter, the writeSortedFile() method allocates a new byte[] of diskWriteBufferSize (default 1 MB) on every invocation. This method is called once per spill and once more for the final file write, so in a spill-heavy shuffle task the same 1 MB array is allocated and discarded repeatedly, adding unnecessary GC pressure.
This PR promotes the write buffer to a lazily-initialized instance field so that it is allocated on the first call to writeSortedFile() and reused across subsequent spills. Lazy initialization ensures that the buffer is not allocated until it is actually needed, so tasks that insert no records (and therefore never call into the write path) pay no extra cost.
Does this PR introduce any user-facing change?
No.
How was this patch tested?
Existing UTs.
Was this patch authored or co-authored using generative AI tooling?
No.