Support for PBS cluster with Singularity and mpirun
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 258
- Forks
- 113
- Avg merge
- 1d 3h
- Merged PRs (30d)
- 5
Description
Hello all, thanks for the amazing library!
I am in the process of porting my code to NeMo 2.0 and NeMo Run. I work with a GH200 HPC cluster and run commands via PBS with Singularity with the NeMo container image and mpirun for going multi-node. Could I use some of the executors to run that? Does the DockerExecutor support Singularity/Apptainer? Would it be possible I make my own executor, and is there a tutorial for that?
I appreciate any help!
EDIT 1: I also have access to enroot 3.5, but haven't used it before.
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 the executor APIs and the DockerExecutor entry point to understand how launch environments are represented. Check whether PBS, Singularity/Apptainer, mpirun, or enroot are addressed elsewhere; done would be documented or implemented support for the requested cluster workflow, including guidance for a custom executor.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, hpc, python
- Domain
- devops, distributed-systems, hpc
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100