Investigating stockpile stewardship applications for world’s largest computer chip

ALBUQUERQUE, N.M. — Sandia National Laboratories and its partners at two other national labs have announced a project to investigate the application of Cerebras Systems‘ Wafer-Scale Engine technology. The immediate target is to accelerate advanced simulation and computing applications in support of the nation’s stockpile stewardship mission.

The National Nuclear Security Administration’s Advanced Simulation and Computing program is sponsoring the work and Sandia, Lawrence Livermore and Los Alamos national labs will collaborate with Cerebras Systems on the project.

“The ultimate goal of NNSA’s Advanced Memory Technology research and development program is to develop technologies for use in future computing system procurements,” said ASC program director Thuc Hoang. “We are funding research in technologies that have the potential to deliver 40 times the application performance of our forthcoming NNSA exascale systems.”

The Cerebras Wafer-Scale Engine, currently the largest computer chip in the world, was built specifically for artificial intelligence and machine learning work, said Andrew Feldman, founder and CEO of Cerebras Systems. “The engine contains 2.6 trillion transistors, 850,000 artificial intelligence cores and powers the Cerebras CS-2, the industry’s fastest artificial intelligence computer,” he said.

Simon Hammond, federal program manager for the ASC’s Computational Systems and Software Environments program, said, “This collaboration with Cerebras Systems has great potential to impact future mission applications by enabling artificial intelligence and machine learning techniques, which are an emerging component of our production simulation workloads.”

The new contract is part of NNSA’s post-Exascale-Computing-Initiative investment portfolio, which has the objective of sustaining the technology research and development momentum, and strong engagement with industry that the initiative had started via its PathForward program. It aims to foster a more robust domestic high-performance computing ecosystem by increasing U.S. industry competitiveness in next-generation high-performance computing technologies.

“We anticipate technologies developed as part of the program will be tested on the Advanced Simulation and Computing program’s advanced architecture prototype systems and will eventually affect the production of advanced and commodity technology platforms used by the three labs,” said Robert Hoekstra, senior manager of the extreme scale computing group at Sandia.

Feldman said his company is proud to have been selected for the work.

“Cerebras is excited to collaborate with the pioneering researchers and scientists at Sandia, Lawrence Livermore and Los Alamos national laboratories,” he said. “Cerebras exists to enable researchers and scientists to push the boundaries of current knowledge, helping them solve problems that are intractable on existing computer infrastructure, as well as vastly accelerate cutting-edge simulation workloads. Our multiyear partnership with the Advanced Simulation and Computing program will expand the boundaries of the application of artificial intelligence and high-performance computing to physics across a range of important applications.”

James H. Laros III, Sandia project lead and Distinguished Member of Technical Staff, said he is looking forward to the collaboration. “The technology holds great potential for impacting how we accomplish our mission in the future.”

Sandia National Laboratories is a multimission laboratory operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration. Sandia Labs has major research and development responsibilities in nuclear deterrence, global security, defense, energy technologies and economic competitiveness, with main facilities in Albuquerque, New Mexico, and Livermore, California.