Supporting NNSA's stockpile stewardship mission through simulation in lieu of underground testing
Beginning of Sierra build

Sierra

Sierra, Livermore’s next advanced technology high performance computing system, joined LLNL’s lineup of supercomputers in 2018. The new system provides computational resources that are essential for nuclear weapon scientists to fulfill the National Nuclear Security Administration’s stockpile stewardship mission through simulation in lieu of underground testing. Advanced Simulation and Computing (ASC) Program scientists and engineers use Sierra to assess the performance of nuclear weapon systems as well as nuclear weapon science and engineering calculations. These calculations are necessary to understand key issues of physics, the knowledge of which later makes its way into the integrated design codes. This work on Sierra has important implications for other global and national challenges such as nonproliferation and counterterrorism.

The IBM-built Sierra supercomputer provides four to six times the sustained performance and five to seven times the workload performance of Sequoia, with a 125 petaFLOP/s peak. At approximately 11 megawatts, Sierra is also about five times more power efficient than Sequoia. By combining two types of processor chips—IBM’s Power 9 processors and NVIDIA’s Volta graphics processing units (GPUs)—Sierra is designed for more efficient overall operations and is a promising architecture for extreme-scale computing. Additional details and user documentation are available.

For Sierra’s installation, the CORAL partnership (Collaboration of Oak Ridge, Argonne, and Livermore) was formed to procure high performance computers from multiple vendors. IBM was selected as the vendor for LLNL. In late 2016, LLNL acquired three small-scale “early access” (EA) versions of Sierra, consisting of IBM Minsky compute nodes with 20 Power 8 cores each and 4 NVIDIA Pascal GPUs. These small systems featured components only one generation behind those of Sierra. The EA systems enabled application porting and tuning in advance of the CORAL Sierra system delivery and acceptance. To enable this work, beta software co-designed by the CORAL laboratories and IBM was installed on the EA systems.

Processor Architecture

IBM Power and NVIDIA Volta processors

Cluster/System Usage

Capability parallel computing

Processor Clock Rate/Process Clock Speed

3.1 GHz

Nodes

4,320

Cores per Node

44

Total Cores

190,080

Memory per Node

256 GB (CPU), 64 GB (GPU)

Total Memory

1.38 PB