AWS ParallelCluster 3.10 is now generally available. Key features of this release include support for Amazon Linux 2023 and Terraform. With Terrafrom support, customers can automate deployment and management of clusters similar to how they use Terraform to automate other parts of their AWS infrastructure. Other important features in this release include:
Support for connecting clusters to an external Slurm database daemon (Slurmdbd) to follow best practices of enabling Slurm accounting in a multi-cluster environment.
A new allocation strategy configuration to allocate EC2 Spot instances from the lowest-priced, highest-capacity availability pools to minimize job interruptions and save costs.
For more details on the release, review the AWS ParallelCluster 3.10.0 release notes.
AWS ParallelCluster is a fully-supported and maintained open-source cluster management tool that enables R&D customers and their IT administrators to operate high-performance computing (HPC) clusters on AWS. AWS ParallelCluster is designed to automatically and securely provision cloud resources into elastically-scaling HPC clusters capable of running scientific, engineering, and machine-learning (ML/AI) workloads at scale on AWS.
Source:: Amazon AWS