Add a new NVTX range for task GPU ownership #11596
Open
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
jihoonson
changed the title
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Outdated suggestions cannot be applied.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.
If you have ever been curious about the semaphore-based GPU concurrency control mechanism, you may have wondered at some point exactly what tasks are holding the semaphore at a given point of time during query processing. This PR can help you in that case. This PR adds a new NVTX range that shows what task owns GPU in the nsys profile result. This feature is off by default (since I don't think it's always useful), and can be enabled by setting
spark.rapids.sql.traceTaskGpuOwnership = true. The screenshot below shows an example nsys result with the new semaphore ranges. The orange boxes withSem-${taskAttemptId}represent the ranges in which each spark task was holding the semaphore.