AMD and Cray unveil what is anticipated to be the world’s quickest supercomputer – defining a brand new customary for high-performance computing and pushing the expertise boundaries of computational science to take the business into the Exascale period.
The new system, referred to as Frontier, is deliberate to come back on-line within the U.S. in 2021 with over 1.5 exaflops of processing energy. The complete system contract award is valued at greater than $600M USD for the system and expertise growth. Together with the Cray Shasta structure, AMD is happy to construct on management {hardware} and software program with the singular imaginative and prescient of fixing the hardest computing challenges on the planet at present. AMD improvements in Frontier embody:
- High Performance Computing (HPC) and Artificial Intelligence (AI) optimized, customized AMD EPYC CPU and purpose-built Radeon Instinct GPU processors
- High Bandwidth Memory (HBM)
- Tightly built-in 4:1 GPU to CPU ratio
- Custom, high-speed coherent Infinity Fabric connection
- Enhanced, open ROCm programming setting for AMD CPUs and GPUs help.
With this announcement, AMD and Cray are driving a brand new HPC paradigm to help the advanced compute, interconnect, software program and storage necessities that Exascale computing calls for.
SANTA CLARA, Calif. — May 7, 2019 — AMD (NASDAQ: AMD) at present joined the U.S. Department of Energy (DOE), Oak Ridge National Laboratory (ORNL) and Cray Inc. in asserting what is anticipated to be the world’s quickest exascale-class supercomputer, scheduled to be delivered to ORNL in 2021. To ship what is anticipated to be greater than 1.5 exaflops of anticipated processing efficiency, the Frontier system is designed to make use of future technology High Performance Computing (HPC) and Artificial Intelligence (AI) optimized, customized AMD EPYC™ CPU, and AMD Radeon™ Instinct GPU processors. Researchers at ORNL will use the Frontier system’s unprecedented computing energy and subsequent technology AI strategies to simulate, mannequin and advance understanding of the interactions underlying the science of climate, sub-atomic buildings, genomics, physics, and different necessary scientific fields.
“AMD is proud to partner with Cray and ORNL to deliver what is expected to be the world’s most powerful supercomputer,” mentioned Forrest Norrod, senior vp and basic supervisor, AMD Datacenter and Embedded Systems Group. “Frontier will feature custom CPU and GPU technology from AMD and represents the latest achievement on a long list of technology innovations AMD has contributed to the Department of Energy exascale programs.”
AMD improvements for use within the Frontier system embody:
- Future-generation High Performance Computing (HPC) and Artificial Intelligence (AI) optimized, customized AMD EPYC CPU, and Radeon Instinct GPU processors supported by High Bandwidth Memory (HBM) and intensive blended precision ops for optimum deep studying efficiency;
- A customized high-bandwidth, low-latency coherent Infinity Fabric, connecting 4 AMD Radeon Instinct GPUs to at least one AMD EPYC CPU per node;
- An enhanced model of the open supply ROCm programming setting, developed with Cray to faucet into the mixed efficiency of AMD CPUs and GPUs.
“We are excited to work with the team at AMD to deliver the Frontier system to Oak Ridge National Laboratory,” mentioned Steve Scott, senior vp and CTO at Cray. “Cray’s Shasta supercomputers are designed to support leading edge processor technologies and high-performance storage, all tightly interconnected by Cray’s new Slingshot network. The combination of Cray and AMD technology in the Frontier system will dramatically enhance performance at scale for AI, analytics, and simulation, enabling DOE to further push the boundaries of scientific discovery.”
AMD has a proud supercomputing historical past and a long-standing engagement with DOE, beginning with the Jaguar supercomputer in 2005 and Titan supercomputer in 2011. The Frontier system leverages years of exascale expertise investments by DOE. The contract award contains expertise growth funding, a middle of excellence, a number of early-delivery techniques, the principle Frontier system and multi-year techniques help.
“Frontier represents the state-of-the art in high-performance computing. Designing and standing up a machine of its scope requires working closely with industry, partnerships which not only enable breakthrough science but also ensure American scientific and economic competitiveness on the global stage,” mentioned Jeff Nichols, affiliate laboratory director for Computing and Computational Sciences, ORNL. “We are delighted to work with AMD to integrate the CPU and GPU technologies that enable this extremely capable accelerated node architecture.”
Additional Resources
- AMD Exascale Computing Technologies
- Cray Shasta Architecture
Follow AMD datacenter developments on Twitter @AMDServer
