Perlmutter (pictured above) will deliver nearly four exaflops of AI performance for more than 8,000 scientists.

On Thursday, the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory, which is DOE Office of Science’s (SC’s) mission high-performance computing facility, launched supercomputer Perlmutter. It is designed for the most demanding, AI-based, “extreme-scale” science applications such as developing new energy sources, improving energy efficiency, discovering new materials, mapping the universe, and analyzing massive data sets from scientific experimental facilities. (Silicon Angle)

Perlmutter

The system is named in honor of Saul Perlmutter, an astrophysicist at Berkeley Lab who shared the 2011 Nobel Prize in Physics for the discovery that the rate at which the universe expands is accelerating.

The HPE Cray system integrates leading-edge technologies with the latest NVIDIA (NASDAQ: NVDA) A100 GPUs, AMD “Milan” EPYC CPUs, a novel HPE Slingshot high-speed network, and a 35 petabyte – all FLASH – scratch file system. The system utilizes 6,159 of the highly advanced Nvidia A100 Tensor Core graphics processing units

Perlmutter’s NVIDIA A100 GPUs with tensor cores are designed with deep learning in mind. Perlmutter’s 35 Petabyte solid-state file system can feed data to the processors at 5 Terabytes per second and a fast, intelligent internal network routes that data where it needs to go quickly and effectively.

The supercomputer is highly energy efficient because its GPUs consume significantly less energy than CPUs. The system is direct liquid-cooled and boasts of a Power Usage Effectiveness (PUE) of 1.05-1.08, among the best in the world.

AI and Supercomputing

NVIDIA CEO Jensen Huang said in a virtual congratulatory message: “Perlmutter’s ability to fuse AI and high-performance computing will lead to breakthroughs in a broad range of fields from materials science and quantum physics to climate projections, biological research and more.”

“People are exploring larger and larger neural-network models and there’s a demand for access to more powerful resources, so Perlmutter with its A100 GPUs, all-flash file system, and streaming data capabilities is well-timed to meet this need for AI,” said Wahid Bhimji, acting lead for NERSC’s data and analytics services group.

Related Story:  Microsoft’s Massive New Supercomputer for AI

Image Credit: Nvidia