Update allreduce benchmark for torch #6271
Conversation
Signed-off-by: Iman Tabrizian <[email protected]>
WalkthroughThe all-reduce microbenchmark script was extensively refactored. The previous TensorRT-based implementation was replaced with a dynamic PyTorch approach using the Changes
Estimated code review effort3 (~45 minutes) Poem
✨ Finishing Touches
🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
|
/bot skip --comment "benchmark script" |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/microbenchmarks/all_reduce.py(3 hunks)
🪛 Ruff (0.12.2)
tests/microbenchmarks/all_reduce.py
60-60: Line too long (144 > 120)
(E501)
157-157: Line too long (155 > 120)
(E501)
🧰 Additional context used
🪛 Ruff (0.12.2)
tests/microbenchmarks/all_reduce.py
60-60: Line too long (144 > 120)
(E501)
157-157: Line too long (155 > 120)
(E501)
🔇 Additional comments (8)
tests/microbenchmarks/all_reduce.py (8)
21-29: LGTM! Imports align with the new PyTorch-based implementation.The new imports properly support the refactored benchmark with fusion modes and CUDA graph capabilities.
32-35: Good addition of CUDA graph support parameter.The function signature changes are appropriate. Starting from size 1 allows for more comprehensive benchmarking of small tensor sizes.
52-58: Well-structured benchmark setup with realistic tensor shapes.The 2D tensor shape
(bs, hidden_size)better reflects real-world ML workloads compared to 1D tensors. The inner/outer loop counts provide good statistical sampling.
59-61: Header properly reflects the new benchmark dimensions.The extended header appropriately includes the new fusion and version columns.
108-114: Benchmark function correctly handles fusion output shapes.The logic properly extracts the tensor from the tuple output when using RESIDUAL_RMS_NORM fusion.
115-149: Excellent timing implementation with proper CUDA graph support.The timing setup includes all necessary components:
- Proper warmup iterations before graph capture
- Separate CUDA stream to prevent interference
- Delay kernel for measurement stability
- Profiler integration for detailed analysis
160-166: Smart size scaling strategy.The logic intelligently scales batch size when hidden size would exceed 4096, maintaining realistic tensor dimensions while achieving the desired total size scaling.
168-182: Command-line argument properly integrated.The new
--enable-cudagraphargument is correctly added and passed to the benchmark function.
|
PR_Github #12611 [ skip ] triggered by Bot |
|
PR_Github #12611 [ skip ] completed with state |
Signed-off-by: Iman Tabrizian <[email protected]>
Summary by CodeRabbit
Refactor
New Features
Bug Fixes
Description
Test Coverage
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.