v0.9 Release

• Support for ParallelCluster 3.0
• Add tag (version) as a parameter for deployment
• Simplify the deployment by removing need for monitoring tarball and monitoring dir. This will default to /home/${cfn_cluster_user}/aws-parallelcluster-monitoring

CustomActions:
        OnNodeConfigured:
          Script: https://raw.githubusercontent.com/aws-samples/aws-parallelcluster-monitoring/main/post-install.sh
          Args:
            - v0.9

add support for P Instances

This release adds the support for P instances to the GPU dashboard.
Thanks @amilsted for the Pr.

GPU support

Initial support for GPUs has been added:

1. nvidia-dcgm container for collecting GPUs metrics
2. Custom GPU Monitoring Dashboard

moving from git clone to wget .. tar.gz

This release enable to safely create new branches by moving from git clone .. to wget .. tar.gz

cost fixed

Cost Dashboard fixed after the release of PC 2.9.

fix CloudWatch logs

thanks to @nonatofabio for finding this bug.

externalize dashboard installation

v0.3-beta

fix the MD with 'bash'

post-install refactoring

Refactoring

1) remove useless yum installations
2) use the zip package from git-hub and fix teh unzip command accordingly

initial release

v0.1-alpha

initial commit