FLC | DOE, HPE unveil plans for El Capitan supercomputer

DOE, HPE unveil plans for El Capitan supercomputer

March 17, 2020

Research at the U.S. Department of Energy’s (DOE) National Nuclear Security Administration (NNSA) will soon be driven by a supercomputer projected to be 10 times faster than today’s most powerful systems, made possible by collaborations with Hewlett Packard Enterprise (HPE) and other industry partners.

The DOE’s Lawrence Livermore National Laboratory (LLNL) has named the new system El Capitan, after the imposing vertical rock formation in nearby Yosemite National Park. Palo Alto, California-based HPE will build the new system, which is expected to be delivered in early 2023. El Capitan will be managed and hosted by LLNL for use by the three NNSA national laboratories: LLNL, Sandia National Laboratories, and Los Alamos National Laboratory.

The system will enable advanced simulation and modeling to support the U.S. nuclear stockpile and ensure its reliability and security. LLNL is managing the new system for the NNSA and has developed techniques for creating faster, more accurate modeling for stockpile modernization and inertial confinement fusion, a key aspect of stockpile stewardship.

LLNL researchers also will use the system to explore new applications that integrate artificial intelligence (AI) and machine learning into high performance computing (HPC) workloads, and are already applying HPE supercomputing and AI solutions to make breakthroughs in medical and drug research initiatives, including:

• Accelerating cancer drug discovery from six years to one year through a partnership with multinational pharmaceutical company GlaxoSmithKline, the National Cancer Institute, and other DOE national laboratories through the ATOM (Accelerating Therapeutics for Opportunities in Medicine) consortium.

• Understanding the dynamics and mutations of proteins that are linked to 30% of human cancers, by collaborating with The National Cancer Institute and other partner institutions.

HPE is optimizing the DOE’s El Capitan to power complex and time-consuming 3D exploratory simulations for NNSA missions that today’s state-of-the-art supercomputers cannot successfully manage. El Capitan will provide opportunities for researchers to explore new applications using emerging, data-intensive workloads such as modeling, simulation, analytics, and AI to support future NNSA missions.

The new supercomputer will be more powerful than the Top 200 fastest supercomputers in the world combined, and more than 30% more powerful than estimates calculated seven months ago. This was made possible by collaborative efforts to combine HPE’s Cray Shasta system and Slingshot interconnect, a specialized HPC networking solution, with next-generation processors and graphics processing units from Advanced Micro Devices Inc. (AMD) of Santa Clara, California.

LLNL and HPE are expanding their partnership to actively explore adding HPE optics technologies, a computing solution that uses light to transmit data, to further improve El Capitan’s power efficiency, reliability and ability to cost-effectively increase global system bandwidth capabilities. HPE’s optics technologies stem from R&D efforts related to PathForward, a program backed by the DOE’s Exascale Computing Project.