Lawrence Livermore National Laboratory announced Wednesday that it has begun installing components for its upcoming El Capitan supercomputer, which is expected to be fully operational by the end of next year. One of the key components of the system is AMD’s upcoming Instinct MI300 accelerated processing unit.
“We have started receiving and installing components for El Capitan, the first #exascale #supercomputer.” Tweet Read by LLNL. “It will still be some time before we deploy for national security purposes in 2024, but we are excited to see our long-standing commitment become a reality.”
Like Frontier and Aurora, El Capitan is built by HP Enterprise as it is based on HPE’s Shasta supercomputer architecture. The machine is expected to deliver over 2 ExaFLOPS of performance when completed in mid-2024, making it the fastest supercomputer in the world at that point.
This supercomputer is based on AMD’s Instinct MI300 hybrid processor with 24 general purpose Zen 4 cores, a CDNA 3 based compute GPU and 128 GB of HBM3 memory. This processor is manufactured by TSMC using one of the nodes belonging to the N5 (5nm class) family.
image 1 of Four
AMD has been testing the Instinct MI300 processor internally for several months now and it looks like they are finally ready to start offering these parts to one of their major customers. Based on the images published by LLNL, engineers in the lab have already installed quite a few servers.
In supercomputer deployments, it’s common to install parts of the system first to make sure everything works at the hardware and software level. So even though El Capitan will be ready for full deployment in his 2024, about a year before everything is complete, it’s time for LLNL to start installing the first parts. El Capitan will be the industry’s first supercomputer to use hybrid processors that pack both general-purpose and stream processors.