Add config to disable columnar shuffle for complex types #2992
+117
−42
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
Member
|
It may be worth enabling |
andygrove
reviewed
Dec 27, 2025
...k/src/main/scala/org/apache/spark/sql/comet/execution/shuffle/CometShuffleExchangeExec.scala
Outdated
Show resolved
Hide resolved
Shekharrajak
force-pushed
the
fix/issue-2904-columnar-shuffle-complex-types
branch
from
00fe67f to
dee4966
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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 #2904
Rationale for this change
Benchmarks show that Comet columnar shuffle with complex/nested types (struct, array, map) is significantly slower than Spark's default shuffle. This PR adds a configuration flag to disable columnar shuffle for complex types, falling back to Spark shuffle for better performance until the underlying performance issues can be addressed.
What changes are included in this PR?
Add spark.comet.columnar.shuffle.complexTypes.enabled config (default: false)
When disabled, queries with complex types in the shuffle schema fall back to Spark shuffle
Update columnarShuffleSupported() to check this config for StructType, ArrayType, and MapTyp
How are these changes tested?
Updated existing columnar shuffle tests for complex types to explicitly enable the new config
Added new test "Fallback to Spark for complex types when config is disabled (default)" that verifies struct, array, and map types fall back to Spark shuffle when the config is disabled