National Energy Research Scientific Computing Center (NERSC) has officially dedicated Perlmutter, which is said to be one of the fastest supercomputers to date for AI calculations. NERSC plans to use Perlmutter to create what will be the largest 3D map of the visible universe to date, probe subatomic interactions for green energy sources, and much more.
The supercomputer 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. It will process dozens of galaxy exposures from one night to know where to point DESI the next night. According to NVIDIA, which accounts for much of Perlmutter’s hardware, the computer will be able to analyze data collected by DESI for years in just a few days that would take weeks or months on prior systems.
The initial version of the Perlmutter supercomputer includes 1536 nodes, each with a 64-core AMD Epyc 7763 processor and four Nvidia A100 GPUs. The result is a system with nearly four exaflops of AI performance for more than 7,000 researchers. The system is also equipped with 35 petabytes of storage space.
Later in 2021, Perlmutter will also be expanded with an additional 3072 CPU-only nodes, each of which will have dual AMD Epyc 7763 chips.
DESI’s map aims to shed light on dark energy, the mysterious physics behind the accelerating expansion of the universe. Dark energy was largely discovered through the 2011 Nobel Prize-winning work of Saul Perlmutter, a still-active astrophysicist at Berkeley Lab who will help dedicate the new supercomputer named for him.
The design of this 3D map is only the first project that the Perlmutter supercomputer is scheduled to take part in. More than two dozen applications are getting ready to be among the first to ride the 6,159 NVIDIA A100 Tensor Core GPUs in Perlmutter, the largest A100-powered system in the world. They aim to advance science in astrophysics, climate science, and more.