search

aws / aws-parallelcluster

AWS ParallelCluster is an AWS supported Open Source cluster management tool to deploy and manage HPC clusters in the AWS cloud.
https://github.com/aws/aws-parallelcluster
Apache License 2.0
828 stars 312 forks source link

PCluster 3 Documentation Missing for Capacity Blocks #6087

Open stefan-maxar opened 8 months ago

stefan-maxar commented 8 months ago

Hello!

I came across the following while attempting to use capacity blocks with PCluster 3.8.0. The current cluster configuration file reference documentation does not list CAPACITY_BLOCK as an option for CapacityType within the Slurm scheduling queue section (see: https://docs.aws.amazon.com/parallelcluster/latest/ug/Scheduling-v3.html#yaml-Scheduling-SlurmQueues-CapacityType).

It is referenced indirectly in the AllocationStrategy section, and is documented properly in the AwsBatchQueues properties section, but not in the SlurmQueues properties section.

hanwen-pcluste commented 8 months ago

We will work on the doc. Thank you!

enrico-usai commented 4 months ago

There is now this useful blog post that can help: https://aws.amazon.com/blogs/hpc/enhancing-ml-workflows-with-aws-parallelcluster-and-amazon-ec2-capacity-blocks-for-ml/