Global supercomputer leader Cray Inc., Seattle WA, announced will evaluate alternative processor design points, including the potential use of 64-bit ARM and other power and performance efficient processing technologies for its supercomputers and data analytics systems.
As part of this effort, Cray was recently awarded a research and development contract from the United States Department of Energy’s Office of Science and the National Nuclear Security Administration under a program called FastForward 2. This contract taps Cray to explore a range of topics around architectures of 64-bit ARM processors for high performance computing.
Additionally, Cray is working with Cavium Inc. to deliver Cray clusters based on Cavium’s 48-core work load optimized ThunderX ARM processors. The goal of this collaboration is to analyze ARM solutions for selected workloads to investigate the ARM value proposition in supercomputing.
“Our Adaptive Supercomputing vision centers on system designs that integrate diverse processing technologies into a unified architecture,” says Steve Scott, Cray senior vice-president and chief technology officer. “We see alternatives such as 64-bit ARM, custom ASICs and low-power Intel processors as enabling technologies for certain HPC and analytics workloads, and a natural fit for our strategy. Along with our current partnership with Intel, our participation in the DOE FastForward 2 project and our work with other technology providers, such as Cavium and ARM, gives us the opportunity to take a closer look.”
“The algorithmic processing performed in HPC environments impacts the lives of billions of people and the planning for Exascale computing presents some significant power challenges to this industry,” adds Mike Muller, chief technology officer, ARM. “The collaboration between ARM and its ecosystem partners has created a proven track record of energy-efficiency leadership and we firmly believe this can be applied to the supercomputing market. ARM welcomes Cray’s pioneering research to explore how 64-bit ARM technology can improve performance density for future supercomputing platforms.”