High-Performance Computing User Facility

NREL's High-Performance Computing (HPC) User Facility features state-of-the-art computational modeling and predictive simulation capabilities to help researchers and industry reduce the risks and uncertainty of adopting new and innovative energy technologies.

Photo of the Eagle supercomputer

The HPC User Facility provides computing capabilities to scientists and engineers working on solving complex computational and data analysis problems related to energy efficiency and renewable energy technologies. The work performed on NREL's HPC systems leads to increased efficiency and reduced costs for these technologies, including wind and solar energy, energy storage, and the large-scale integration of renewables into the electric grid.

Kestrel Computing System

Kestrel is NREL's latest HPC system dedicated to advancing renewable energy and energy efficiency technologies. During its installation in the HPC User Facility, Kestrel will complement NREL's current supercomputer, Eagle. When its transition is completed—in early 2023—Kestrel will accelerate energy efficiency and renewable energy research at a pace and scale more than five times greater than Eagle, with approximately 44 petaflops of computing power. See the Kestrel Versus Eagle Supercomputer Stack-Up infographic.

Eagle Computing System

The Eagle system is a Linux cluster that uses a fast InfiniBand network. It is composed of 2,114 interconnected "compute nodes" with 4,228 Intel Skylake processors and a total of 76,104 cores—along with 14 petabytes of high-speed data storage. The peak performance of Eagle is approximately 8 petaflops, or 8 million billion floating point (mathematical) operations per second. 

Eagle was designed and built by Hewlett-Packard Enterprise and has an innovative warm water liquid-cooling system that allows waste heat to be captured for reuse. 

Mass Storage System

The Amazon Web Services Mass Storage System is an additional data archival resource available to those who use NREL's HPC computing systems.


Share