Sign up for our newsletter and get the latest HPC news and analysis.

Best Practices Selecting an HPC Cluster

Wyoming Supercomputing CenterEverything from life sciences to the financial industry are relying on HPC clusters to perform complex and critical operations. Moving forward, there will be more data, more applications and a lot more reliance on various HPC systems. So the all-important question comes in – How do you select, deploy and manage it all? Fortunately, IBM, Intel and NCAR have teamed up to explain their view on best practices selecting an HPC cluster using the process behind building the NCAR Wyoming Supercomputing Center.

As new research and engineering environments are expanded to include more powerful computers to run increasingly complex computer simulations, the management of these heterogeneous computing environments continues to increase in complexity as well. Integrated solutions that include the Intel Many Integrated Cores (MIC) architecture can dramatically boost aggregate performance for highly-parallel applications.

In this presentation, NCAR will reveal some of their plans for “Yellowstone” – one of the largest x86-based systems in N. America – located at the NCAR-Wyoming Supercomputer Center and officially opened in October 2012. NCAR Distinguished Software Engineer, Rory Kelly, will describe the selection criteria and best practices in deploying a system of this magnitude.

In addition, Intel Parallel Programming Senior Engineer and published author, James Reinders, will highlight the technology attributes and benefits of Intel Xeon Phi™ and how this new coprocessor contributes a significant performance impact and positive return on investment for customers.

Join this great event to get details on System x integrated solutions! Furthermore, you’ll learn how the IBM Platform Computing approach to managing distributed workloads across platforms utilizes a powerful methodology that will “change the game” for cluster administrators and end users.

Remember, as you build out and manage your own HPC cluster – It will be critical to align resources, workloads and utilization to the needs of both your business and the end-user. Your capabilities and efficiency with management can potentially directly dictate just how quickly you can quantify key pieces of information.

watch now

Resource Links: