-
Notifications
You must be signed in to change notification settings - Fork 230
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix Databricks 13.3 BroadcastHashJoin using executor side broadcast fed by ColumnarToRow [Databricks] #10230
Merged
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
Note, not kicking the build because I know the databricks one is currently broken from other issues |
build |
revans2
approved these changes
Jan 22, 2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
But I really would like more eyes on it than just mine.
jlowe
approved these changes
Jan 22, 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Outdated suggestions cannot be applied.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 #10165
Here the issue is that Databricks supports a special executor side broadcast in BroadcastHashJoin and in this case the BroadcastHashJoin ended up being on the CPU. The BroadcastHashJoin can't be fed by a ColumnarToRow. It expects to be fed by an exchange that specifically handles going to a single partition.
There is a similar issue in Databricks 12.2 that we fixed by inserting an extra GpuColumnarToRow and then an Exchange. I use similar logic here to fix this and filed a followup to investigate the performance of now having 2 exchanges (#10229).
The beginning plan looks like:
After this fix the plan looks like:
Added an integration test for databricks. The test throws an exception without the fix and then passes with the fix. This new logic only occurs on Databricks 13.3 and the parameters to AQEShuffleReadExec changed between 12.2 and 13.3 so the shim code went into a specific Databricks 13.3 file.