AWS recently introduced many HPC services that boost the performance and scalability of Computational Fluid Dynamics (CFD) workloads on AWS. These services include: Amazon FSx for Lustre, Elastic Fabric Adapter (EFA), and AWS ParallelCluster 2.5.1. In this technical post, we will first explore these three services. Then we will describe the steps involved in using AWS ParallelCluster to set up an HPC system with EFA and Amazon FSx Lustre to run a CFD workload. The CFD application that you will set up during this blog post is Simcenter STAR-CCM+ – a popular CFD application from Siemens.
AWS and Siemens are dedicated to enhancing the customer experience of running Simcenter STAR-CCM+ apps on AWS. In this technical post, we will cover the steps and the best practices for running Simcenter STAR-CCM+.
The Simcenter STAR-CCM+ application runs on an HPC cluster. This cluster is optimized with EFA and Amazon FSx for Lustre — all of which are managed by AWS ParallelCluster. AWS ParallelCluster simplifies the deployment process to such an extent that you can set up your HPC cluster with a high throughput parallel file system (Amazon FSx for Lustre), a high-throughput and low-latency network interface (EFA), and high-bandwidth network inter-connects (100 Gbps using C5n instances) in less than 15 minutes.
The high level overview of the process is
- Creating an HPC infrastructure stack on AWS, which will include:
- Setting up AWS ParallelCluster to launch your cluster
- Enabling EFA
- Setting up Amazon FSx Lustre file system
- Connecting to a remote desktop session using NICE DCV
- Installing the Simcenter STAR-CCM+ application and how to submit a Simcenter STAR-CCM+ job to HPC cluster.
Once the application is set up, we run a LeMans 104M cell benchmark. Here are sample scaling results for the LeMans 104M cell benchmark case. The results indicate how well the simulation scales with EFA-based networking and the powerful Amazon EC2 C5n instances.
You can read the entire article and follow along step by step here