[Question]: Will NCCL ext-profiler supports to enable both profiler and inspector meanwhile?
Open
Nobody has claimed this yet.
question
- Dominant language
- C++
- Stars
- 5.1k
- Forks
- 1.4k
- Avg merge
- 2h 3m
- Merged PRs (30d)
- 2
Description
Question
The variable NCCL_PROFILER_PLUGIN now can only points to profiler plugin(ext-profiler/example) or inspector plugin(ext-profiler/inspector/example).
Any plan to merge these two plugins so that both coll events and metrics can be traced during job training.
Looking forward for your response, thanks.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading ext-profiler/example and ext-profiler/inspector/example, then trace how NCCL_PROFILER_PLUGIN selects one plugin. Determine the design needed to trace both collective events and metrics during training. Done means the combined behavior is supported and its configuration is clear.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- observability, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100