deepspeedai / deepspeedai/DeepSpeed
[BUG] Autotuner is not launching experiments with correct hostfile setting
Open
Nobody has claimed this yet.
bug
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
It seems like when the autotuned is spawning experiments it is not passing the correct --hostfile parameter to the experiment commands what leads to failure for all jobs that do not use default location for the hostfile.
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
Locate the autotuner entry point that spawns experiment commands and trace how the --hostfile setting is constructed or passed. Reproduce the issue with a non-default hostfile; done means autotuned experiments receive that setting and no longer fail for that configuration.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100