NVIDIA has announced at ISC 2024 that nine new supercomputers globally are using NVIDIA Grace Hopper™ Superchips. These superchips help with faster scientific research and discovery. The systems deliver 200 exaflops, or 200 quintillion calculations per second, of energy-efficient AI processing power.
Also Read: What are the Top Skills Needed for Digital Transformation?
About the Grace Hopper Installation
New Grace Hopper architecture-based supercomputers are planned to be deployed worldwide to enhance advanced computation power across research domains.
The French Alternative Energies and Atomic Energy Commission (CEA) and Eviden—an Atos Group firm—announced the delivery of the supercomputer EXA1-HE. This system is based on Eviden’s BullSequana XH3000 technology, integrating a patented warm-water cooling system. EXA1-HE is a 477-node system, all of them based on Grace Hopper architecture.
Among others are Helios at the Academic Computer Centre Cyfronet, Krakow, Poland, provided by Hewlett Packard Enterprise (HPE); Alps at the Swiss National Supercomputing Centre at ETH Zurich, also from HPE; JUPITER at Jülich Supercomputing Centre in Germany; DeltaAI at the National Center for Supercomputing Applications at the University of Illinois Urbana-Champaign; and Miyabi at the Joint Center for Advanced High-Performance Computing, Japan.
These supercomputers mark a milestone in computational research capacity with Grace Hopper architecture’s processing power and efficiency.
“AI is accelerating research into climate change, speeding drug discovery and leading to breakthroughs in dozens of other fields,” said Ian Buck, vice president of hyperscale and HPC at NVIDIA. “NVIDIA Grace Hopper-powered systems are becoming an essential part of HPC for their ability to transform industries while driving better energy efficiency.”
Growing Presence of NVIDIA Supercomputers with Grace Hopper
Isambard-AI and Isambard, 3 from the University of Bristol in the U.K. and systems at the Los Alamos National Laboratory and the Texas Advanced Computing Center in the U.S., join a growing wave of NVIDIA Arm-based supercomputers using Grace CPU Superchips and the Grace Hopper platform.
Also Read: How Security Orchestration Automation and Response (SOAR) Streamlines Incident Response?
Advancing Sovereign AI
Countries worldwide are doing even more to pursue even more efficient AI-based supercomputers, recognizing the strategic importance of sovereign AI. This includes investments in locally owned and hosted data, infrastructure and workforces that bring innovation.
The GH200, powered by the Arm-based NVIDIA Grace CPU and NVIDIA Hopper GPU architectures with NVLink-C2C interconnect technology, drives scientific supercomputing centers worldwide. Many are trying to go from system installation to real scientific applications in a matter of months, not years.
The first phase of Isambard-AI comprises an HPE Cray EX2500 supercomputer installed with 168 NVIDIA GH200 Superchips, making it extraordinarily efficient. When the remaining 5,280 NVIDIA Grace Hopper Superchips arrive at the University of Bristol’s National Composites Centre this summer, performance will be expected to increase by around 32 times.
Driving Scientific Breakthroughs
NVIDIA’s accelerated computing platform brings together NVIDIA Hopper architecture-based GPUs, NVIDIA Grace CPU Superchips, NVIDIA Grace Hopper Superchips, NVIDIA Quantum-2 InfiniBand networking, and a complete range of NVIDIA AI and HPC software.
FAQs
1. What is NVIDIA Grace Hopper?
The NVIDIA GH200 Grace Hopper™ Superchip is a breakthrough processor designed from the ground up for giant-scale AI and high-performance computing (HPC).
2. How does NVIDIA Grace Hopper™ Superchips speed scientific research and discovery?
Grace Hopper™ Superchips offer extreme performance for running high-fidelity and scalable quantum simulations on accelerators and seamlessly interacting with future quantum hardware infrastructure.
3. How does the NVIDIA Grace Hopper Superchip contribute to scientific advancements?
The NVIDIA Grace Hopper Superchip accelerates research into various fields, including climate change, drug discovery, and numerous other scientific domains. Its integration into supercomputers enhances computational efficiency and drives innovation across industries.
4. What is “Sovereign AI,” and how does it relate to NVIDIA’s Grace Hopper Superchips?
Sovereign AI refers to the strategic initiative undertaken by countries to invest in domestically owned and hosted data, infrastructure, and workforces to drive innovation. NVIDIA’s Grace Hopper Superchips play a vital role in constructing more efficient AI-based supercomputers, aligning with the objectives of Sovereign AI initiatives.
[To share your insights with us as part of editorial or sponsored content, please write to sghosh@martechseries.com
