NVIDIA Powers Over 400 of the World’s 500 Fastest Supercomputers

NVIDIA technologies power more than 400 of the world's 500 fastest supercomputers, according to the latest TOP500 rankings released at the ISC High Performance conference in Hamburg, Germany [1]. The company's reach now spans 81% of the list, a gain of 17 systems from the previous ranking [1]. The momentum is strongest among new deployments: 90% of systems entering the list are built on NVIDIA technologies [1]. The company's GPUs accelerate a record 238 systems, while NVIDIA networking connects 376 of the TOP500 machines, the majority via Quantum InfiniBand [1]. NVIDIA DGX systems, which typically house 4 to 8 GPU modules on an independent system board, have been featured in TOP500 supercomputers as the company expanded its datacenter footprint [6]. Adoption of the NVIDIA Grace CPU reached 26 systems, up eight from the previous list, with nearly 2.5 million Grace CPUs shipped [1]. The Grace Hopper Superchip pairs an NVIDIA GPU with the Grace CPU, allowing shared memory with minimal overhead [1]. The top eight systems on the Green500 efficiency ranking run on NVIDIA GPUs, and nine of the top 10 use NVIDIA technologies [1]. KAIROS, a Grace Hopper system at France's University of Toulouse, leads the Green500 at 73.3 gigaflops per watt [1]. A record 35 NVIDIA AI HPC supercomputers are in development across Europe, equipping more than 3 million researchers with next-generation infrastructure [1]. Among them is JUPITER at Germany's Jülich Supercomputing Centre, Europe's fastest supercomputer and its first to reach exascale, which is mapping the human brain at cellular scale and simulating Earth's climate [1]. The buildout extends globally, from a new AI factory in South Africa to national AI systems in Saudi Arabia, Singapore, and Vietnam [1]. The newest systems entering the list run on the NVIDIA Blackwell architecture, with B200 and GB200 systems debuting across Asia, Europe, and the U.S., and the first GB200 systems appearing in Japan [1]. The TOP500, updated twice yearly, ranks the world's fastest supercomputers, while the Green500 measures computational efficiency per watt [1]. NVIDIA's trajectory in high-performance computing follows decades of architecture development, from the Turing microarchitecture introduced with the GeForce RTX 20 series in 2018 — the first consumer product line to implement hardware-enabled real-time ray tracing — through to the datacenter-focused Grace and Blackwell platforms [2].