The Future of HPC

It’s a busy time in Austin, Texas. We’ve just wrapped up Dell World, we’re in the middle of the last quarter of the year, and the Supercomputing 2015 conference is here. All the leaders in supercomputing have come to Austin for barbeque, bands, and bright ideas – and Dell will be in the front lines, talking about our vision for removing complexity and risk and empowering world-changing innovation and discovery in this enormously important area of the computing landscape.

Perhaps you don’t know, but Dell has a good track record with high-performance computing (HPC). At the moment, according to the Top 500 list, we have clusters at the University of Texas, Cambridge University, the University of Warsaw, Lawrence Livermore Laboratories, and Saudi Aramco, just to name a few. And, as you’ve heard in the news lately, we’re transforming Dell into an enterprise powerhouse that will extend and accelerate our HPC capabilities in domain-specific applications, optimized platforms, and the full range of solutions and services – stretching from desktop to petascale.

Why? Because HPC is becoming more and more critical for solving fundamental human problems. Data sets are growing exponentially, opportunities and needs for analyzing and using it emerge every day across the globe, and our customers in the public and private sectors welcome every opportunity to drive change they can find. HPC has become a foundational requirement for science and business, and our customers know that they can rely on us for HPC insight and innovation.

But there are always new opportunities to accelerate change, and we think we’ve found a few.

One of the traditional challenges of HPC has been comparing different approaches and vendors. Each vendor has offered different ways of proving themselves, demonstrating performance, and conducting testing and reporting. We don’t think that’s right.

Why? Because enterprises value the benefits that come from the freedom to choose. Organizations deserve a level playing field that helps them:

  • Control their path to success,
  • Know and manage their costs, and
  • Future-proof their business.

That’s what Dell has always offered — but now we’re going beyond anyone’s expectations. This week we announced the Dell HPC System Portfolio, our way to advance and democratize high-performance computing by combining the flexibility of custom systems with the simplicity and value of pre-configured solutions. We also announced the expansion of the Austin-based Dell HPC Innovation Lab, and continued support for the next generation of HPC technologies under development by Intel, not only by being the first OEM to join Intel’s Fabric Builders program but also with the Dell Networking H-Series switches and adapters featuring the Intel Omni-Path Architecture.

But we’re not just about products — we’re here to drive a visionary approach for the future of HPC. On November 12, we united with Intel and other industry leaders to form the OpenHPC Collaborative Project. This project will provide a new, open source framework to support the world’s most sophisticated HPC environments. Made up of both vendors and their customers, it offers new opportunities and possibilities within the world of supercomputing. The project aims to deliver a stable environment for testing and validation, a robust and diverse open source software stack, and a flexible, heterogeneous framework for configuration. These new developments will result in better agility, stronger consistency, and a level playing field that delivers reduced costs.

This is the right move. It’s directly in line with our vision and, most importantly, delivers what the marketplace demands and needs. Proprietary, expensive, limited approaches to computing are becoming passé, and HPC is no different.

Organizations require simple and inexpensive, agile and scalable over monolithic and expensive. That’s why open APIs matter, why industry-standard technologies matter, and why interoperability matters. Basically, the more inflexible, proprietary, and complicated something is, the less valuable it is to the marketplace – because it gets in the way of the flexibility that enables success, growth, and innovation.

At Dell, we’re on the front lines of making HPC matter. We’re looking forward to the day when, through the OpenHPC Collaborative Project, all vendors have a consistent, open source software stack, standard methods for testing and validation, the ability to use heterogeneous components together, and the capacity for giving customers reduced costs. We certainly don’t tolerate any approach that keeps enterprises from making the choices that are right for them – and these announcements prove it.

To put it simply, we enable enterprises with HPC technologies that are simple, efficient, and work together well for better control, insight, and long-term value. We’re proud to bring you new approaches, ideas, and opportunities — and we’re glad to help you by accelerating HPC that’s open, efficient, and optimized for value.

To learn more about our HPC capabilities visit our HPC homepage, follow us on Twitter, or contact your Dell representative.

Glenn Keels

About the Author: Glenn Keels