Skip to main content

AWS HPC

High Performance Computing

Run your HPC workloads at virtually any scale

Why HPC on AWS?

Run your large, complex simulations and deep learning workloads in the cloud with a complete suite of high performance computing (HPC) products and services on AWS. Gain insights faster, and quickly move from idea to market with virtually unlimited compute capacity, a high-performance file system, and high-throughput networking.

Benefits

    Access a broad range of cloud-based services, like machine learning (ML) and analytics, plus HPC tools and infrastructure to quickly design and test new products.

    Gain on-demand access to compute capacity. Skip the wait and save more time to focus on solving complex problems without worrying about cost and infrastructure constraints.

    Find the perfect fit for your infrastructure needs and solve real-world business problems with your choice of the largest selection and capacity of HPC services, fast networking, and storage on AWS.

Explore HPC on AWS

AWS offers fully managed services, infrastructure building blocks, and open-source tooling for high performance computing. Find what you need to do, then go to the service that provides it. You can start with a fully managed service, then add the compute, networking, and storage building blocks your workloads need.

What you need to do
Common uses
AWS service
Run and manage HPC clusters

Simulation, modeling, and AI/ML workloads across engineering and the sciences

AWS Parallel Computing Service
Run large batches of jobs

High-throughput and containerized pipelines for genomics, financial analysis, and rendering

AWS Batch
Give teams remote desktops

Engineering and research desktops for preparing inputs and reviewing results

Amazon WorkSpaces
Stream graphics-intensive apps

Engineering, scientific, and visualization applications delivered to any device

Amazon AppStream 2.0
Run on GPUs and accelerators

AI and ML training, and accelerated simulation, seismic, and CFD

AWS EC2 Accelerated Computing
Connect nodes with fast networking

Tightly coupled simulation like CFD and weather, and distributed model training

Elastic Fabric Adapter
Add high-performance shared storage

Scratch and shared data for running simulation, genomics, and ML jobs

Amazon FSx
Store and stage large datasets

Input datasets, simulation output, research data lakes, and long-term archives

Amazon S3
Visualize results remotely

3D simulation and engineering results, without moving data to the desktop

Amazon DCV
Self-manage your own clusters

Simulation, modeling, and research on clusters you configure and control

AWS ParallelCluster
Run a research portal

Virtual desktops and interactive applications for research and engineering teams

Research and Engineering Studio on AWS

Use cases

Tightly coupled workloads

Run tightly coupled parallel MPI applications at virtually any scale, including computational fluid dynamics (CFD), computer-aided engineering (CAE), molecular dynamics, weather and climate modeling, and seismic and reservoir simulation.

High-throughput computing and loosely coupled workloads

Run large numbers of independent jobs across many instances, powering electronic design automation (EDA), Monte Carlo and financial risk simulations, image processing, and genomics analysis.

Accelerated computing

Decrease time to results for diverse workloads using GPUs, FPGAs, and Amazon custom silicon such as AWS Trainium and AWS Inferentia, for workloads ranging from AI and machine learning model training to scientific and engineering modeling, protein structure prediction, and Cryo-EM.

Interactive workflows

Run human-in-the-loop workflows: prepare inputs, run simulations, and visualize and analyze results in real time, then use what you learn to refine the next run.

AWS High Performance Computing Competency Partners

AWS High Performance Computing Competency Partners help customers accelerate their digital innovation in the areas of HPC spanning high performance solvers, high throughput computing, HPC workload management, and foundational HPC technology. AWS HPC Competency Partners offer agile, flexible, and cost-effective technology offerings for HPC on AWS supporting genomics, computational chemistry, seismic processing, computer aided engineering, as well as emerging applications, such as deep learning and autonomous driving.

Find a partner