Add support for NCCL
Open
@shi-eric is already working on this.
Since Jan 29, 2025.
- Dominant language
- Python
- Stars
- 7.1k
- Forks
- 624
- Avg merge
- 3d 17h
- Merged PRs (30d)
- 5
Description
- All five collective communication functions should be supported: https://docs.nvidia.com/deeplearning/nccl/user-guide/docs/api/colls.html
ncclSendandncclRecvshould also be supported: https://docs.nvidia.com/deeplearning/nccl/user-guide/docs/api/p2p.html- NCCL can be statically linked or dynamically linked, including from conda-forge (https://anaconda.org/conda-forge/nccl)
- Binary size is an obvious concern, so by default dynamically linked NCCL will be assumed.
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.
Assessment
This issue has not been assessed yet.