[WIP]: Add nvtx annotations to cudf-polars shuffle - #22718
Draft
TomAugspurger wants to merge 1 commit into
Draft
Conversation
This adds nvtx annotations for shuffles in cudf. Note that we don't use `nvtx.annotate` which is push/pop based on the current thread. This implies some kind of partent-child relationship. But if we have two Shuffle actors, it would be inappropriate to say that the second one to start is a child of the first. So instead we use `nvtx.start_range` and `nvtx.end_range`.
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
github-actions
Bot
added
Python
3 tasks
TomAugspurger
added
improvement
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
This adds nvtx annotations for shuffles in cudf. Note that we don't use
nvtx.annotatewhich is push/pop based on the current thread. This implies some kind of partent-child relationship. But if we have two Shuffle actors, it would be inappropriate to say that the second one to start is a child of the first.So instead we use
nvtx.start_rangeandnvtx.end_range.Joins will often involve multiple shuffle (e.g. left and right). We set the nvtx
categoryto thecollective_idthat uniquely identifies the shuffleNote: I'm using https://gist.github.com/TomAugspurger/cebf449f7621f69ff70a5452f1ca5706 as a small benchmark that forces shuffle-based joins for a cudf-polars query.