How can you improve the performance of your HPC (High Performance Computing) workload, if the bottlenecks are shared disk access and latency between nodes?

Ways to improve HPC performance

  • Amazon FSx for Lustre provides scalable high-performance shared storage.
  • Elastic Fabric Adapter (EFA) is a high-performance network interface for EC2 instances.
  • Cluster Placement group will provision EC2 instances which are physically close to each other.