To be clear, it’s not that Cray and Livermore aren’t disclosing the chip suppliers these decisions haven’t been made yet. “The El Capitan system procurement was written in such a way that the Livermore team in working with us at Cray can make a late-binding decision on the node architecture choice there’s lots of different options and that part of the market is changing very rapidly between the different CPUs and GPUs that are available,” said Cray CEO Pete Ungaro at a media briefing held yesterday, announcing the signing of the contract. Notably, the chip and accelerator component suppliers have not been announced. Įl Capitan’s Shasta system will be comprised of “fat” CPU-GPU nodes, utilizing Cray’s Slingshot interconnect and a future generation of ClusterStor storage.
The announcement marks the second award announcement to come out of the CORAL-2 program, the joint DOE-NNSA effort to procure up to three exascale supercomputers with a potential budget of $1.8 billion.
Compared with its geological namesake, the famous “El Cap” in Yosemite National Park, Livermore’s El Capitan compute blades when laid end to end would scale the peak of El Capitan more than three times, said Cray. With a peak speed of more than 1.5 exaflops, El Capitan will be based on Cray’s Shasta architecture and provide advanced capabilities for modeling, simulation and artificial intelligence (AI).
SMOOZE FOR EL CAPITAN FULL
The contract, valued at $600 million, calls for Cray to deliver El Capitan to Livermore in late 2022 with full production targeted for late 2023, enabling NNSA to continue to perform essential functions for the United States’ Nuclear Stockpile Stewardship Program. Since 1987 - Covering the Fastest Computers in the World and the People Who Run ThemĬray has won the bid to build the first exascale supercomputer for the National Nuclear Security Administration (NNSA) and Lawrence Livermore National Laboratory (LLNL).
0 kommentar(er)
