Ensure an executor broadcast is in a single batch [databricks] #10660
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
… of RequireSingleBatch Signed-off-by: Navin Kumar <[email protected]>
Signed-off-by: Navin Kumar <[email protected]>
NVnavkumar
changed the title
|
build |
jlowe
reviewed
Apr 4, 2024
...main/spark330db/scala/org/apache/spark/sql/rapids/execution/GpuExecutorBroadcastHelper.scala
Outdated
Show resolved
Hide resolved
...c/main/spark330db/scala/org/apache/spark/sql/rapids/execution/GpuBroadcastHashJoinExec.scala
Outdated
Show resolved
Hide resolved
Signed-off-by: Navin Kumar <[email protected]>
Signed-off-by: Navin Kumar <[email protected]>
|
build |
jlowe
reviewed
Apr 5, 2024
...main/spark330db/scala/org/apache/spark/sql/rapids/execution/GpuExecutorBroadcastHelper.scala
Outdated
Show resolved
Hide resolved
...main/spark330db/scala/org/apache/spark/sql/rapids/execution/GpuExecutorBroadcastHelper.scala
Outdated
Show resolved
Hide resolved
Signed-off-by: Navin Kumar <[email protected]>
NVnavkumar
requested review from
revans2,
tgravescs,
GaryShen2008,
NvTimLiu and
gerashegalov
as code owners
|
build |
Comment on lines
+325
to
+327
| """ | ||
| select /*+ BROADCAST(df_school) */ * from df, df_school where df.id == df_school.id | ||
| """ |
There was a problem hiding this comment.
nit: PySpark API looks nicer and would not require resorting to view registration along the lines
df.join(broadcast(df_school), df.id == df_school.id) There was a problem hiding this comment.
Understood, but this test requires a join hint to ensure a broadcast hash join is used in this test.
There was a problem hiding this comment.
Signed-off-by: Navin Kumar <[email protected]>
|
build |
jlowe
approved these changes
Apr 12, 2024
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.
Fixes #10645.
This wraps the GpuShuffleCoalesceIterator into a GpuCoalesceIterator with a coalesce goal of RequireSingleBatch. This ensures that we always get a single batch to use for the broadcast join. Also added an integration test that tweaks the batch size to test this codepath.