AWS HPC Blog

Category: AWS ParallelCluster

Recent improvement to Open MPI AllReduce and the impact to application performance

Recent improvement to Open MPI AllReduce and the impact to application performance

Our team engineered some Open MPI optimizations for EFA to enhance performance of HPC codes running in the cloud. By improving MPI_AllReduce they improved scaling – matching commercial MPIs. Tests show gains for apps including Code Saturne and OpenFOAM on both Arm64 and x86 instances. Check out how these tweaks can speed up your HPC workloads in the cloud.

Securing HPC on AWS – isolated clusters

Securing HPC on AWS – isolated clusters

This post references AWS ParallelCluster. Check out AWS Parallel Computing Service (AWS PCS), our new managed Slurm service for running HPC and AI workloads on AWS. In this post, we’ll share two ways customers can operate HPC workloads using AWS ParallelCluster while completely isolated from the Internet. ParallelCluster supports many different network configurations to support […]