by Alan Gara, exascale chief architect
High Performance Computing, the science of aggregating computing power to solve complex science and engineering problems, is a powerful set of emerging technologies with tremendous economic and strategic implications for companies and entire nations. It has applications in weather mapping, energy use and much more. Today, many nations are rapidly scaling up their investments in HPC with an eye on economic competitiveness, scientific leadership and other benefits that come from these critical capabilities. While it’s exciting to see progress being made in this field around the world, we also applaud the United States government’s commitment to promoting continued US leadership in HPC.
This commitment is especially important today, as we celebrate the one year anniversary of the the National Strategic Computing Initiative (NSCI). Established by the White House Office of Science and Technology Policy, the NSCI was created to sustain and enhance US scientific, technological, and economic leadership position in HPC research, development, and deployment.
As a key participant in the NSCI, we have worked to support the program’s goals, and have made significant progress. For example, we have collaborated with the broad HPC ecosystem under the Department of Energy’s CORAL program, creating a standard architecture for HPC systems based on our Scalable System Framework that can easily adapt to meet different end-user requirements and budgets. We’ve also been a part of the DesignForward research and development program, accelerating the development of the next generation of supercomputers based on Intel’s X86 architecture. And we’ve ramped up research in the exciting areas of neuromorphic and quantum computing, both of which have the potential to be truly revolutionary.
It has been a productive and rewarding first year. But there is still much more to do. We need to continue to drive government, commercial and academic research and investment across all the areas that enable continued development. These include process technology, fabric architecture and development, optical components, and architectural innovation. Just as important will be to enable common software tools, programming methodologies and libraries and toolkits that are optimized to the system hardware.
Accomplishing this will help us to make sure that HPC solutions meet the broadest possible range of needs, ranging from new users for whom ease of implementation and use are critical, to the most demanding and experienced users for whom raw performance is paramount. Robust, common infrastructure and software layers will also enable users to migrate across different solutions, and ensure the flexibility and configurability to design systems based on specific solution needs, whether that is real-time analytics, modeling and simulation of problems that scale across thousands of nodes, or new applications still unforeseen.
We believe that NSCI will continue to play an important role in bringing this vision to life. We hope the US Department of Energy, one of the lead agencies driving NSCI, will continue supporting long-term investments and partnerships with vendors that enable closer coupling of HPC component R&D with overall system design, via programs such as APEX and the Exascale Initiative. And we look forward to the strengthening of mechanisms to deal with the uncertainties that are always part of long-term, speculative R&D projects in emerging technology areas.
Innovation and discovery are difficult to predict, but this is one of the most important initiatives in computing, and we’re excited to continue being a part of it.