[graph_trainer] Add benchmark.py for forward-backward profiling#2805
Closed
SherlockNoMad wants to merge 2 commits intogh/SherlockNoMad/10/basefrom
Closed
[graph_trainer] Add benchmark.py for forward-backward profiling#2805SherlockNoMad wants to merge 2 commits intogh/SherlockNoMad/10/basefrom
SherlockNoMad wants to merge 2 commits intogh/SherlockNoMad/10/basefrom
Conversation
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. [ghstack-poisoned]
SherlockNoMad
requested review from
aditvenk,
tianyu-l,
xmfan and
yiming0416
as code owners
SherlockNoMad
added a commit
that referenced
this pull request
Apr 3, 2026
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ghstack-source-id: d854c7c Pull Request resolved: #2805
meta-cla
bot
added
the
CLA Signed
…iling" Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ``` ================================================== Benchmark Results (mode=aot_fx_trace) ================================================== Steps: 10 (after 3 warmup) Mean step time: 910.11 ms Tokens/step: 8192 Peak memory: 42.40 GiB (reserved) TFLOPS: 260.65 MFU: 26.35% ================================================== ``` [ghstack-poisoned]
SherlockNoMad
added a commit
that referenced
this pull request
Apr 3, 2026
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ghstack-source-id: a808b3e Pull Request resolved: #2805
This was referenced Apr 3, 2026
SherlockNoMad
added a commit
that referenced
this pull request
Apr 6, 2026
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ghstack-source-id: a808b3e Pull Request resolved: #2805
SherlockNoMad
added a commit
that referenced
this pull request
Apr 6, 2026
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ghstack-source-id: a808b3e Pull Request resolved: #2805
SherlockNoMad
added a commit
that referenced
this pull request
Apr 7, 2026
Add a lightweight benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional single-step torch profiler trace capture. Update CLAUDE.md with benchmark and profiling instructions. ghstack-source-id: a808b3e Pull Request resolved: #2805
SherlockNoMad
added a commit
that referenced
this pull request
Apr 9, 2026
Add a benchmark script that measures forward_backward_step performance (eager vs aot_fx_trace) without running the optimizer. Reports mean step time, peak GPU memory, TFLOPS, and MFU. Supports optional torch profiler trace capture. Update CLAUDE.md with Benchmark and Profiling sections documenting usage. Port of PR #2805.
4 tasks
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
1 participant
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.
Stack from ghstack (oldest at bottom):
Add a lightweight benchmark script that measures forward_backward_step
performance (eager vs aot_fx_trace) without running the optimizer. Reports
mean step time, peak GPU memory, TFLOPS, and MFU. Includes optional
single-step torch profiler trace capture.
Update CLAUDE.md with benchmark and profiling instructions.