kubeflow / mpi-operator

Kubernetes Operator for MPI-based applications (distributed training, HPC, etc.)
https://www.kubeflow.org/docs/components/training/mpi/
Apache License 2.0
417 stars 209 forks source link

Update stale examples #596

Open jarulsamy opened 8 months ago

jarulsamy commented 8 months ago

Hi,

Some of the examples are no longer working. For example, pi-mpich.yaml refers to docker image tags that no longer exist. In addition, it would be really nice if we had some larger examples (similar to the tensorflow ones) that don't need a GPU.

If there are no objections, I am happy to get a PR put together for this myself.

terrytangyuan commented 8 months ago

Feel free to submit PRs!

tenzen-y commented 8 months ago

Regarding pi-mpich.yaml, the image isn't outdated. Although we release the example images every release, the image hasn't yet been released since the MPICH implementation was contributed recently. This means that we don't have any releases for the MPICH implementation.