GpuShuffleCoalesceIterator acquire semaphore after host concat #4396
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
Signed-off-by: Alessandro Bellina <abellina@nvidia.com>
jlowe
reviewed
Dec 20, 2021
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuShuffleCoalesceExec.scala
Show resolved
Hide resolved
jlowe
approved these changes
Dec 20, 2021
|
build |
…rf/join_sem_tweaks
|
build |
abellina
added a commit
to abellina/spark-rapids
that referenced
this pull request
Dec 30, 2021
…A#4396) * GpuShuffleCoalesceIterator acquire semaphore after host concat Signed-off-by: Alessandro Bellina <abellina@nvidia.com> * Add semaphore acquire for batches without columns
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Outdated suggestions cannot be applied.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.
Closes #4395
This is a small optimization that was spotted while looking into NDS Q64 traces. With the change, Q64 can save up to 3 seconds (though it changes quite a bit from run to run). When executing this over all of NDS, it saved ~1 minute for the whole run chipping away at times a few hundred ms, up to 5 seconds for q94.
I saw some queries being slower, with the worst case being q42 (which for 1 sample out of 10 was 2x slower). I have not been able to reproduce this case, with all subsequent runs at 1x or above. This was a 3.8 second in the last weekly run, with the patch it's hovering between 3.6 and 4.5.