The latest renowned supercomputer could help solve some of the most important questions in astrophysics. VentureBeat reports that the National Energy Research Scientific Computing Center has officially dedicated Perlmutter, considered one of the fastest supercomputers for AI, and it will start by helping build the largest 3D map of the visible universe to study dark energy accelerating the cosmos’ expansion.
The machine will process the data from the dark energy spectroscopic instrument to guide the observations. Perlmutter will process “dozens” of galaxy exposures from a given nocturnal survey to dictate where DESI goes next.
As AMD notes, the initial release of Perlmutter features 1,536 nodes each with a 64-core Epyc 7763 processor and four NVIDIA A100 GPUs. The result is a system with nearly four performance exaflops (plus 35 petabytes of storage) when handling AI tasks, and significantly reduced compute times. Perlmutter could process a year of DESI information in “days” instead of weeks or months, NVIDIA said.
A second phase slated for later in 2021 will add 3,072 CPU nodes only with two Epyc 7763 chips.
The map isn’t the only project on the deck. Perlmutter will also study atomic interactions to promote clean energy technologies like biofuels, for example. The map of the universe is ambitious, however, and illustrates the extent to which supercomputers have come to tackle large-scale work.
All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through any of these links, we may earn an affiliate commission.