AWS ParallelCluster v2.9.0
We're excited to announce the release of AWS ParallelCluster Node 2.9.0
This is associated with AWS ParallelCluster v2.9.0
ENHANCEMENTS
- Add support for multiple queues and multiple instance types feature with the Slurm scheduler.
- Replace the previously available scaling components with:
clustermgtd
daemon that takes care of handling compute fleet management operations, included the processing of health check failures coming from EC2 and cluster start/stop operations;slurm_resume
andslurm_suspend
scripts that integrate with Slurm power saving plugin;computemgtd
daemon that monitors the health of the system from the compute nodes. - Replace Auto Scaling Group with plain EC2 APIs when provisioning cluster nodes.
- Register cluster nodes in a Route53 private hosted zone when DNS resolution is enabled for the cluster.
- Register mapping between Slurm node names and EC2 instances in DynamoDB table.
- Create log files for the new components in
/var/log/parallelcluster/
dir.
- Replace the previously available scaling components with: