Improve HPC workloads on AWS for environmental sustainability
HPC Blog
This article provides guidance on optimizing HPC workloads on AWS for environmental sustainability. It covers strategies to reduce the carbon footprint of compute-intensive simulations while maintaining productivity goals.
Specifically, the article covers:
- Choosing the right EC2 instance types and AWS Regions for your workload based on performance and sustainability goals
- Leveraging AWS Graviton instances and purchase options like Spot and Savings Plans for cost and energy efficiency
- Optimizing storage with lifecycle policies and services like FSx for Lustre and S3
- Using low-latency networking like Elastic Fabric Adapter (EFA) and Cluster Placement Groups
- Orchestrating compute resources with AWS Batch and AWS ParallelCluster
- Using remote visualization with NICE DCV for graphics-intensive pre/post-processing
- Evaluating sustainability improvements with proxy measures like cost per job and resource utilization
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Jun 24
2024
2024
A library of HPC Applications Best Practices on AWS
Aug 28
2024
2024
Announcing AWS Parallel Computing Service to run HPC workloads at virtually any scale
Jun 24
2025
2025
Optimize HPC Workloads with the Updated AWS Well-Architected Lens
Jan 5
2026
2026
Where energy meets efficiency: Modernize HPC workflows with Leostream on AWS
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.