Servers: The Right Compute
cancel
Showing results for 
Search instead for 
Did you mean: 

Are you ready for supercomputing?

Bill_Seidle

Discover why it's a good time to consider the promise of supercomputing and assess the impact and evolution of supercomputing applications from the public to the private sector.

HPE HPC - supercomputing - blog.jpgWe live in fast-changing times, filled with technological breakthroughs. Of course, the only ones we hear about are those with broad appeal, like curing a disease or some new app for our smartphones. But in the computing world, something with potentially far-reaching implications has been brewing below the radar: a long-running competition among world governments to build the first exascale computer, one capable of a quintillion (a billion billion, or 1018) calculations per second, or exaFLOPS.

If this technology evolves like others, we should expect these supercomputers to migrate even more into private enterprises. With this transformation around the corner, it is a good time to consider the promise of supercomputing, its applications, its contributions, and its impact as it becomes affordable and available to the free market.

Understanding HPC and supercomputing

A variant of supercomputing, known as high-performance computing (HPC), is very much present and commercially available today. While the terms HPC and supercomputing are often used interchangeably—both use parallel processing of high-end components to expedite the running of advanced applications—there is a distinction, chiefly in performance characteristics. HPC systems are typically capable of a trillion calculations per second (teraFLOPS). Supercomputers operate a thousand times faster (petaFLOPS), making them the world’s fastest computers. As you can imagine, they are prohibitively expensive and large, costing millions of dollars and filling entire rooms.

Applications for supercomputing

Currently, HPC and supercomputing make their greatest impact in scientific research applications, such as developing advanced simulations and models. To understand the practicality of this, consider automobile safety. Before 3D computer modeling, the only way to collect this data was by repeatedly crashing real cars rigged with sensors, which is both expensive and time consuming. Now engineers conduct their safety experiments on virtual vehicles, where they can easily simulate any crash scenario, collect data, tweak variables, and retest.

In medicine, it once took the most powerful computer a whole day to sequence a full human genome for the treatment of rare diseases. With HPC, it now takes about 20 minutes. True supercomputers will render this in a few seconds or less.

In the private sector, however, HPC adoption has been slower but steady. It has been gradually making inroads into the enterprise data center as a hybrid cloud platform and to support big data generators like artificial intelligence (AI), high-performance data analytics (HPDA), and the Internet of Things (IoT).

As these technologies develop, the increasing amount of data moving across networks and clouds will create a demand for more powerful computing resources. In fact, recent trends in HPC server revenue from Hewlett Packard Enterprise (HPE) reflect strong growth in the demand for HPC solutions. In preparation for this uptick, HPE collaborates with providers of leading-edge technologies to build, validate, and deliver secure production-level HPC solutions.

The manifestation of supercomputing

The reason we see supercomputing predominantly in government agencies is because they are closest to the source of the money. The Exascale Computing Project (ECP) is headed by the U.S. Department of Energy’s science and national security laboratories, so it’s no surprise that federally funded research interests, such as energy and security, reap the first benefits of their supercomputing research.

When software becomes available, look for supercomputing to spark breakthroughs in quantum physics, medicine and pharmaceuticals, climate studies, energy development, cosmology, and beyond. Supercomputing has the potential to increase our understanding of nearly every facet of the physical world.

In the private sector, look for more practical consumer applications of this technology to drive economic competition in the free market. Advancements in AI, machine learning, HPDA, and cybersecurity will enable businesses to deliver unprecedented applications and services. For example, with data generated from IoT, entire industries will be able to anticipate, read, and react to the needs of customers even before those needs manifest themselves. It will be a brave new world of consumerism where things show up at your doorstep before you know you need them!

The path forward

Because the spirit of entrepreneurship is free to mingle with supercomputer modeling, look for it to ignite innovations in manufacturing, aerospace, oil and gas, drug production, finance, computing, insurance, healthcare, and other industries. Inevitably, it will set new standards and expectations in the consumer marketplace. In the future, a business won’t be able to compete without access to supercomputer-driven applications.

Getting there depends on IT vendors developing mainstream supercomputing systems with stable architectures that are compatible with existing, industry-standard technologies. But first, they will need to address the new challenges that supercomputing brings, like power delivery, cooling, upgradability, and easy access to your average business.

Companies like HPE, along with its global partners, are already putting this capability within reach via cloud-based supercomputing. Additionally, innovations in liquid-cooling and power delivery are being made that will facilitate growth in supercomputing, AI, and deep learning, building a bridge to the next generation of exascale supercomputers.

Ready to find out more? HPE has an unparalleled portfolio of supercomputing solutions to explore.


Bill Seidle
Hewlett Packard Enterprise

twitter.com/HPE_Servers
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/servers

0 Kudos
About the Author

Bill_Seidle

As the manager of HPC and AI Portfolio Marketing within the Hybrid IT Group at HPE, I lead a worldwide marketing team that develops our HPC and AI differentiated messaging, competitive positioning and value proposition, creates best-in-class supporting assets and digital content and amplifies our compelling message leveraging a variety of web, social, digital and traditional media formats.