Open
Description
Similar to NCCL tests for Kubernetes https://github.com/aws-samples/awsome-distributed-training/tree/main/micro-benchmarks/nccl-tests/kubernetes - it would be great if there was a similar test for NCCOM https://github.com/aws-samples/awsome-distributed-training/tree/main/micro-benchmarks/nccom-tests
- Would this require a new Docker image similar to
public.ecr.aws/hpc-cloud/nccl-tests:latest
-public.ecr.aws/hpc-cloud/nccom-tests:latest
?