Nvidia will create a new supercomputer which is claimed to be the fastest in the world for AI workloads at the National Energy Research Scientific Computer Center (NERSC) in California, reports VentureBeat. The supercomputer will initially be used for building the most ambitious 3D map of the universe by utilizing the 6,000+ Nvidia A100 Tensor Core GPUs onboard. It is named after astrophysicist Saul Perlmutter.
Presently, Perlmutter is delivering almost four exaFLOPS of AI performance, which Nvidia says makes it “the fastest system on the planet on 16 and 32-bit mixed-precision math AI uses”. Its performance will be increased further in “phase two” when the second tranche of CPU cores will be introduced. A “phase two” is due later in 2021 will add 3,072 CPU-only nodes that each have dual Epyc 7763 chips.
Nvidia’s New Supercomputer – Perlmutter
As per Nvidia, more than 7,000 researchers will be working on the Perlmutter system and tens of applications have been developed for achieving the advancement of astrophysics and climate science. The initial version of Perlmutter has 1,536 nodes that each have a 64-core Epyc 7763 processor and four Nvidia A100 GPUs. Nvidia claimed that Perlmutter could process a Dark Energy Spectroscopic Instrument information in just a few days to help in assembling the largest 3D map of the universe visible to date.
“In one project, the supercomputer will help assemble the largest 3D map of the universe visible to date. It will process data from the Dark Energy Spectroscopic Instrument (DESI), a kind of cosmic camera that can capture as many as 5,000 galaxies in a single exposure,” says Dion Harris, Nvidia HPC & AI Product Marketing Lea
“Researchers need the speed of Perlmutter’s GPUs to capture dozens of exposures from one night to know where to point DESI the next night. Preparing a year’s worth of data for publication would take weeks or months on prior systems, but Perlmutter should help them accomplish the task in as little as a few days.”
It is expected that once a 3D map of the universe is completed, it would help researchers to learn more about dark energy, the force behind the ever-accelerating expansion of the universe. In 2011, Saul Perlmutter earned the Nobel Prize for the discovery of dark energy.
Perlmutter machine is expected to help in other projects with similar ambitious goals. Many would use its unique qualities of the Tensor Cores in the A100 GPUs to simulate interactions between atoms, which is considered next to impossible. Perlmutter could be proved a change in the field of materials science, by uncovering ways to develop more efficient batteries, biofuels, and the like.