Memory profiler #2061
Open
Memory profiler #2061
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2061
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
jainapurva
added
topic: performance
There was a problem hiding this comment.
Pull Request Overview
This PR introduces memory profiling capabilities to the microbenchmarking framework, enabling CUDA memory profiling and visualization alongside the standard profiling functionality. Key changes include the implementation of memory profiling functions in profiler.py, updates to configuration files and utility classes to support new profiling options, and the addition of comprehensive tests to validate the functionality.
Reviewed Changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| benchmarks/microbenchmarks/utils.py | Added new fields and updated to_dict to carry memory profiling info |
| benchmarks/microbenchmarks/test/test_benchmark_profiler.py | Added tests for memory profile generation and visualization |
| benchmarks/microbenchmarks/test/benchmark_config.yml | Updated YAML config to include enable_memory_profiler flags |
| benchmarks/microbenchmarks/profiler.py | Added generate_memory_profile and visualize_memory_profile functions, including pickle file validation |
| benchmarks/microbenchmarks/benchmark_inference.py | Integrated memory profiling functions to run alongside model profiler |
| benchmarks/microbenchmarks/README.md | Documented new memory profiling options and usage |
Co-authored-by: Copilot <[email protected]>
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.
This pull request introduces memory profiling capabilities to the microbenchmarking framework, alongside enhancements to the existing profiling infrastructure. Key changes include the addition of a memory profiler, updates to configuration files to enable memory profiling, and new test cases to validate the functionality.
Memory Profiling Enhancements:
generate_memory_profileandvisualize_memory_profilefunctions inbenchmarks/microbenchmarks/profiler.pyto enable CUDA memory profiling and visualization. This includes memory snapshots, stats collection, and HTML visualizations of memory usage._validate_pickle_fileinprofiler.pyto ensure memory profile files are valid and readable.Configuration Updates:
benchmarks/microbenchmarks/README.mdandbenchmarks/microbenchmarks/test/benchmark_config.ymlto includeenable_memory_profileras a configuration option. This allows users to enable memory profiling for specific models.Benchmarking Framework Changes:
benchmark_inference.pyto run memory profiling whenenable_memory_profileris set. Results include memory stats and visualization paths.enable_memory_profilertoBenchmarkConfigand memory profiling-related fields (memory_profile_path,memory_visualization_path,memory_stats) toBenchmarkResultinutils.py.These changes enhance the profiling capabilities of the benchmarking framework, providing deeper insights into memory usage during inference, especially for CUDA-enabled devices.Add support for memory profiler