The US Department of Energy (DOE) announced that Lawrence Berkeley National Laboratory’s National Energy Research Scientific Computing (NERSC) Center has signed a contract with Cray for NERSC’s next-generation supercomputer, a pre-exascale machine slated to be delivered in 2020.

Named “Perlmutter,” in honor of Berkeley Lab’s Nobel Prize winning astrophysicist Saul Perlmutter, it is the first NERSC system specifically designed to meet the needs of large-scale simulations as well as data analysis from experimental and observational facilities.

This new supercomputer has a total contract value of $146 million, including multiple years of service and support, and will more than triple the computational power currently available at NERSC.

It represents DOE Office of Science’s commitment to extreme-scale science, developing new energy sources, improving energy efficiency, and discovering new materials. In addition, the new system has a number of innovative capabilities that will facilitate analyzing massive data sets from scientific experimental facilities, a growing challenge for scientists across multiple disciplines.

We are very excited about the Perlmutter system. It will provide a significant increase in capability for our users and a platform to continue transitioning our very broad workload to energy efficient architectures. The system is optimized for science, and we will collaborate with Cray, NVIDIA and AMD to ensure that Perlmutter meets the computational and data needs of our users. We are also launching a major power and cooling upgrade in Berkeley Lab’s Shyh Wang Hall, home to NERSC, to prepare the facility for Perlmutter.

—NERSC Director Sudip Dosanjh

The new supercomputer will be a heterogeneous system comprising both CPU-only and GPU-accelerated cabinets. It will include a number of innovations designed to meet the diverse computational and data analysis needs of NERSC’s user base and speed their scientific productivity. That includes a new Cray system interconnect, code-named Slingshot, that is designed for data-centric computing; as well as NVIDIA GPUs with new Tensor Core technology, direct liquid cooling, and an all-flash scratch filesystem which will move data at a rate of more than 4 terabytes/sec.