deepspeedai / deepspeedai/DeepSpeed
Resume from Zero OOM，because this line let every SINGLE DP process load ALL optimizer states, Memory Needed * N
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
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 with the linked line in deepspeed/runtime/engine.py at line 1796 and trace the resume-from-zero path around it. Determine how optimizer states are loaded by each data-parallel process, then verify that the completed behavior avoids loading all states into every process without changing resume correctness.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100