Add Python CUDA reduction benchmark with cuda.cccl#10
Open
AnshSinghSonkhia wants to merge 1 commit intoashvardanian:mainfrom
Open
Add Python CUDA reduction benchmark with cuda.cccl#10AnshSinghSonkhia wants to merge 1 commit intoashvardanian:mainfrom
AnshSinghSonkhia wants to merge 1 commit intoashvardanian:mainfrom
Conversation
Introduces reduce_bench.py, a Python script to benchmark parallel reductions on NVIDIA GPUs using the cuda.cccl library, and updates the README with usage instructions and example output. This allows users to compare naive CuPy reductions with optimized CUDA JIT reductions from Python.
Author
|
Hi @ashvardanian, Could you please review this PR? |
Owner
|
Hey @AnshSinghSonkhia! Thanks for the PR! It's a good start! I'll find some time to explore the CCCL Python functionality in more depth and integrate the PR over the next few weeks. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Introduces reduce_bench.py, a Python script to benchmark parallel reductions on NVIDIA GPUs using the cuda.cccl library, and updates the README with usage instructions and example output. This allows users to compare naive CuPy reductions with optimized CUDA JIT reductions from Python.
This solves #9