Jeffrey Preston Bezos Executive Chairman of Amazon | Amazon
Amazon Web Services, Inc. (AWS), a subsidiary of Amazon.com, Inc., has announced the general availability of AWS Parallel Computing Service. This new managed service is designed to facilitate the setup and management of high-performance computing (HPC) clusters, enabling customers to run scientific and engineering workloads at scale on AWS.
The service integrates Amazon Elastic Compute Cloud (Amazon EC2) instances with low-latency networking and storage optimized for HPC workloads. It allows scientists and engineers to quickly scale simulations for model validation while providing system administrators with tools like Slurm, an open-source HPC workload manager, to build and maintain clusters.
"AWS Parallel Computing Service aims to make every scientist and engineer using it the most productive in their field," said Ian Colle, director of advanced compute and simulation at AWS. "Our goal is to provide top-tier HPC capabilities at any scale."
Among the early adopters are Marvel Fusion, Maxar Intelligence, RONIN, and The National Renewable Energy Laboratory (NREL). Moritz von der Linden, CEO of Marvel Fusion, highlighted the service's ability to deliver highly available and easy-to-upgrade HPC cluster management capabilities. Travis Hartman from Maxar Intelligence noted its potential for better cluster visibility and compute provisioning.
Nathan Albrighton from RONIN emphasized that the service simplifies building and operating HPC environments using APIs. Michael Bartlett from NREL pointed out that AWS Parallel Computing Service could improve research efficiency by reducing overhead through automated updates and observability management features.
AWS has a history of supporting HPC workloads with services like AWS ParallelCluster, AWS Batch, Elastic Fabric Adapter, Amazon FSx for Lustre high-performance storage, and dedicated AMD-, Intel-, and Graviton-based HPC compute instances. The new service builds on this foundation by offering a fully-managed solution that reduces operational burdens.
The service is now available in multiple regions including US East (Ohio), US East (N. Virginia), US West (Oregon), Europe (Frankfurt), Europe (Stockholm), Europe (Ireland), Asia Pacific (Sydney), Asia Pacific (Singapore), Asia Pacific (Tokyo).
For more information about AWS Parallel Computing Service or to get started with setting up an HPC cluster using Slurm on AWS infrastructure, visit https://aws.amazon.com/pcs/.
---
