Draft
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
-rank-0and current vLLMrank0torch profiler filenamesasync_llmcoordinator traces from per-rank gap analysis discoveryProblem
With
vllm 0.17.0, torch profiler outputs rank traces like:dp0_pp0_tp0_dcp0_ep0_rank0....pt.trace.json.gzdp0_pp0_tp1_dcp0_ep1_rank1....pt.trace.json.gzand also emits a coordinator trace like:
...async_llm....pt.trace.json.gzGapAnalyzer.detect_trace_files()only matched the older-rank-0style. That meant rank detection failed, the code fell back to*.json.gz, andasync_llmwas incorrectly counted as an extra rank.Validation
Validated against a real
vllm 0.17.0Kimi K2.5 trace directory from an MI355 run:async_llmrank0..rank3)Notes
I did not add a unit test in this patch because I was validating against the live trace artifacts first. The next clean step would be a small test covering both filename styles and the
async_llmexclusion.