Expanding your HPC cluster capacity with Red Hat Enterprise Linux for HPC in Azure

High performance computing (HPC) has a long history of unlocking incredible scientific discovery around the world as well as powering private industry through highly parallel, cluster-based supercomputing. Today, many organizations from private enterprises to global government labs rely on the power of HPC to do product development, physical simulation, and scientific research.

Red Hat Enterprise Linux (RHEL) has been an invaluable platform to many of these organizations in their HPC journey as the premiere operating system. The argument for using RHEL for HPC is clear: enhanced security capabilities, greater stability and a trusted ecosystem of hardware and software solutions that make HPC possible.

We often hear from customers that the cloud is important in their HPC workflow to expand their on-premises capacity in order to finish their HPC jobs in the time that’s needed to meet their organization’s goals. For that reason, Red Hat is excited to announce that it's bringing RHEL for HPC to Microsoft Azure, making it easier to expand HPC cluster capacity to the cloud. This offering includes:

• A RHEL image designed to help HPC customers get up and running as quickly as possible in Azure
• Simplified, self-service cloud purchasing to allow HPC customers to burst into the cloud at the moment they need it

An HPC specific image to quickly deploy

With RHEL for HPC in Azure, customers will have a RHEL image that they can use to rapidly deploy in the cloud. Specifically, this is a base RHEL image with an embedded Ansible system role. Red Hat worked closely with Microsoft to identify a pre-selected set of installed software, tools and configurations that can make running HPC workloads on Azure performant and quick to deploy.

The Ansible system role can install important libraries, drivers and tools that make running an HPC workload on Azure performant. Some of the software and configurations the system role can install through automation are:

• MPI libraries
• Hardware accelerator drivers, tuned where applicable
• HPC diagnostic scripts
• Monitoring tools
• Health checks
• Kernel rules
• HPC utilities

This RHEL image is only available when selecting RHEL for HPC in Azure Marketplace. Upon selecting that offering, a customer will simply need to run the system role which then makes the RHEL image ready for running the job.

A new simplified way to purchase

Customers who already run their on-premises HPC clusters with RHEL are familiar with RHEL’s existing head node and compute node structured offering. This existing pattern has helped customers to grow their clusters to a massive scale.

When customers choose to run RHEL for HPC in Azure, they will be purchasing a single item with a single price and can do it all self-service through Azure, offering a more seamless purchasing experience and enabling customers to accelerate HPC workloads in the cloud.

Why use the cloud for HPC?

One of the reasons an organization may leverage the cloud for their HPC workloads is to improve time-to-market. Contrary to conventional perception, HPC workloads do often affect an organization’s time-to-market, and therefore time-to-revenue. This is particularly true in enterprise HPC instances, where critical simulations such as those running in the aerospace or automotive sectors have a direct impact on product delivery dates. For these types of use-cases, bursting into the cloud to access additional HPC infrastructure can create tremendous business value for an organization.

From crash test simulation for the automotive industry to computational fluid dynamics in aerospace, RHEL for HPC is a preferred platform for many of the world’s most complex scientific computing problems.

Click here to be notified when this new offering will be available in early 2025.