NVIDIA GB200 and NVIDIA Quantum-X800 InfiniBand Networking Boost Japan’s Research in Life Sciences, Materials Science, Climate and Weather Forecasting, Manufacturing, Laboratory Automation and Quantum Applications
News Summary:
- Two new RIKEN supercomputers for scientific AI and quantum computing, powered by NVIDIA GB200 methods, will place Japan as a driving drive behind AI for science, next-generation industrial analysis and quantum computing.
- RIKEN’s AI and quantum methods will function a complete of two,140 NVIDIA Blackwell GPUs, supporting Japan’s sovereign AI technique and increasing safe home infrastructure for management in science, trade and know-how.
SC25—NVIDIA at the moment introduced that RIKEN, Japan’s main nationwide analysis institute, is integrating NVIDIA GB200 NVL4 methods with two new supercomputers in Japan — one constructed for AI for science and the opposite for quantum computing.
The first system will deploy 1,600 NVIDIA Blackwell GPUs, utilizing the GB200 NVL4 platform and interconnected by NVIDIA Quantum-X800 InfiniBand networking, as a part of RIKEN’s AI for science initiative. The system will advance analysis in areas similar to life sciences, supplies science, local weather and climate forecasting, manufacturing and laboratory automations.
The second system, devoted to quantum computing, will function 540 NVIDIA Blackwell GPUs — additionally utilizing the GB200 NVL4 platform and interconnected by NVIDIA Quantum-X800 InfiniBand networking — to speed up analysis in quantum algorithms, hybrid simulation and quantum-classical computing strategies.
“RIKEN has long been one of the world’s great scientific institutions, and today it stands at the forefront of a new era in computing,” mentioned Ian Buck, vp of hyperscale and high-performance computing (HPC) at NVIDIA. “Together, we’re helping Japan build the foundation for sovereign innovation that will drive breakthroughs to solve the world’s most complex scientific and industrial challenges.”
“Integrating the NVIDIA GB200 NVL4 accelerated computing platform with our next-generation supercomputers represents a pivotal advancement for Japan’s science infrastructure,” mentioned Satoshi Matsuoka, director of the RIKEN Center for Computational Science. “Our partnership will create one of the world’s leading unified platforms for AI, quantum and high-performance computing, allowing researchers to unlock and accelerate discoveries in fields ranging from basic sciences to industrial applications for businesses and society.”
Expanding Partnership With RIKEN
The two new RIKEN methods observe the announcement in August that launched a collaboration between Fujitsu and NVIDIA to codesign a flagship supercomputer with the event code identify FugakuNEXT, the successor to the world-renowned Fugaku supercomputer. The two new GPU-accelerated supercomputers can even be used as proxy machines — platforms for codesigning and creating numerous {hardware}, software program and purposes for FugakuNEXT.
The FugakuNEXT system is deliberate to function FUJITSU-MONAKA-X CPUs, which could be paired with NVIDIA applied sciences utilizing NVIDIA NVLink™ Fusion, new silicon enabling high-bandwidth connections between Fujitsu’s CPUs and NVIDIA’s structure.
FugakuNEXT is predicted to ship 100x better utility efficiency in contrast with supercomputers based mostly on CPUs or different current methods — and will combine production-level quantum computer systems sooner or later.
By combining MONAKA-X and NVIDIA’s newest GPUs, FugakuNEXT will assist form the way forward for scientific discovery by means of innovation in HPC, AI, quantum and their mixtures.
This rising partnership with RIKEN displays Japan’s dedication to innovation and NVIDIA’s assist in bolstering the nation’s computational infrastructure and capabilities in supercomputing and AI for science.
Supercomputing Software Unlocks Scientific Advancements
NVIDIA is already working with RIKEN to develop floating level emulation software program that faucets into NVIDIA Tensor Core GPU efficiency for accelerating conventional scientific computing. This know-how will enable purposes to harness the total energy of GPUs for AI and HPC at RIKEN and supercomputing facilities worldwide.
RIKEN additionally plans to make use of NVIDIA CUDA-X™ — which supplies 400+ extremely optimized GPU-accelerated libraries, microservices and instruments — to spice up its cutting-edge HPC purposes with GPU platforms, serving to advance AI for science and quantum computing initiatives in Japan.
The two new supercomputers will probably be operational in spring 2026, whereas FugakuNEXT is aimed for operation by 2030.