nicolaven
commented
2 years ago Hi there,
In general the code that you find on https://github.com/aws-samples is not supposed to be used in production but instead as a sample code, snippets, you use your projects.
About to your questions:
-
This particular project is constantly updated, maintained, and is also used by a number of customer in production environments. AWS does not commit on supporting it but we typically fix problem or respond to features requests in an "affordable effort" manner.
-
if you are looking for examples of how to use a Slurm REST API, you can have a look at this: https://aws.amazon.com/blogs/hpc/using-the-slurm-rest-api-to-integrate-with-distributed-architectures-on-aws/
-
yes it does support GPUs see here: https://github.com/aws-samples/1click-hpc/blob/main/parallelcluster/config.us-east-1.sample.yaml#L165 in particular, the GPU instances can be uses for compute jobs or for running (3D) interactive applications (using Nice DCV).
-
No there is no API specifically developed for this project but most of the component/services used by this project do have an API: a. ParallelCluster 3.X has an API https://docs.aws.amazon.com/parallelcluster/latest/ug/api-reference-v3.html b. Nice EnginFrame has a REST API (as well as a wsdl web services interface): https://dn3uclhgxk1jt.cloudfront.net/enginframe/packages/2020.1/documentation/EFAdminGuide2020.1.pdf
All,
Stumbled on this while struggling to get slurmrestd set up on pcluster. It looks like this provides a lot of friendly wrappers for HPC type problems. Is this code production-ready? Does it support GPU instances? Is there an API provided?