Built-in GPU performance monitoring during benchmarks#35
Open
KaunilD wants to merge 3 commits intoNVIDIA:mainfrom
Open
Built-in GPU performance monitoring during benchmarks#35KaunilD wants to merge 3 commits intoNVIDIA:mainfrom
KaunilD wants to merge 3 commits intoNVIDIA:mainfrom
Conversation
…tion Starts one gweperf process per worker node before the benchmark and stops it cleanly (SIGINT + 30s timeout fallback to SIGKILL) after. Adds MonitoringConfig/MonitoringFeaturesConfig schema and gweperf_path cluster setting. Monitoring failures are non-fatal; benchmark continues unaffected. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Removes the external gweperf dependency and srtslurm.yaml gweperf_path requirement. Adds src/srtctl/monitor/perfmon.py — a ~90-line script that polls nvidia-smi and writes per-second CSV samples + aggregate JSON on SIGINT. Drops MonitoringFeaturesConfig; MonitoringConfig is now just enabled + sample_interval. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The head node runs GPU workers in most topologies, so excluding it from perfmon left a gap in power/utilization metrics. Remove the head-node filter so all worker nodes get monitored. Made-with: Cursor
KaunilD
force-pushed
the
kdhruv/gweperf_integration
branch
from
e36a925 to
cf68041
Compare
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #35 +/- ##
=======================================
Coverage ? 62.39%
=======================================
Files ? 52
Lines ? 5053
Branches ? 0
=======================================
Hits ? 3153
Misses ? 1900
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Collaborator
Author
uploading graphs created with the collected data. aligned using the timestamp from benchmark.out
|
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.
Summary
perfmon.py) that polls nvidia-smi during benchmarks, writing per-node CSV time-series (util, memory, power, temp) and aggregate JSON summariesmodel.pathpoints to a hub cache root with symlinked snapshotsTest plan
tests/test_monitoring.py— 19 tests passingperf_samples_{node}.csvappears for all nodes including head