Lawrence Livermore National Laboratory announced Monday that the El Capitan exascale supercomputer has been benchmarked as the fastest computing system in the world. The system was evaluated based on the High Performance Linpack standard, which is used by the TOP500 organization to evaluate supercomputing performance.
Table of Contents
Performance Rating
El Capitan earned an HPL score of 1.742 exaFLOPs with a total peak performance of 2.79 exaFLOPs. In comparison, Sierra, LLNL’s previous most powerful system, had a peak performance of 125 petaFLOPs. El Capitan’s peak performance exceeds that of Sierra’s more than 20-fold.
Commenting on the accomplishment, LLNL Lab Director Kim Budil said it “is a testament to the Laboratory’s leadership in driving scientific discovery. It continues a legacy of supercomputing excellence that spans more than 70 years.”
“El Capitan’s extraordinary computing capabilities will allow us to tackle complex challenges that were previously out of reach. We are proud to lead this achievement in partnership with industry, and advance science in ways that will benefit society and the nation as a whole,” Budil added.
System Specs
LLNL’s new supercomputer is built on the Cray Supercomputing EX system by Hewlett Packard Enterprise. The system is also powered by AMD’s Instinct MI300A accelerated processing units, which work to deliver computational performance required by artificial intelligence workloads.
Functions of El Capitan
El Capitan will be used by the National Nuclear Security Administration Tri-Labs — LLNL, Los Alamos National Laboratory and Sandia National Laboratory — to support the Stockpile Stewardship Program and other nuclear security missions. The supercomputer will help with the various tasks to be carried out by Tri-Labs scientists, including weapon performance and safety modeling, high-energy-density physics experiment modeling and AI-based workflows like those involving material discovery, advanced manufacturing and digital twinning.