Is there a way to combine data parallel and model parallel?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
I would like to inquire about the possibility of combining data parallelism and model parallelism in the context of training llm. I found that the model parallel only support 1 batch while data parallel can not distribute one model to many cards. If I have 1000 1080ti cards and I want train a 65B model in a big batch size, what should I do?
Contributor guide
No contributing guide indexed for this repository
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
The issue names no files, tests, or entry points; begin by clarifying whether a concrete FastChat change is wanted for combining data and model parallelism. A complete issue should identify the relevant training path and a test or reproducible configuration, plus define what successful large-batch 65B training looks like.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100