The RIKEN Center for Computational Science, has begun operating a new GPU-based supercomputer called ROQUO, designed to support research that combines quantum computing with HPC. Installed at RIKEN’s facility in Kobe, Japan, ROQUO was developed under the Japanese government’s JHPC-quantum project, which focuses on hybrid platforms that integrate quantum computers and supercomputers. The system was built primarily by DTS Corporation and has now entered operational service.
According to a RIKEN press release, ROQUO is among the first systems in full-scale operation globally to use the GB200 NVL4 architecture. The configuration is designed for scientific and technical computing workloads while also supporting AI applications. Unlike NVIDIA’s larger GB200 NVL72 systems, which are optimized for large-scale AI training and inference, the GB200 NVL4 emphasizes a balance of HPC performance, operational flexibility, and cost efficiency.
Mitsuhisa Sato, Director of the Quantum-HPC Hybrid Platform Division, RIKEN Center for Computational Science, said, “The JHPC-quantum project aims to advance hybrid computing technologies that integrate quantum computers with high-performance computing (HPC) systems. The integration of GPU-accelerated computing and AI technologies has attracted significant attention, and the deployment of this new GPU supercomputer comes at a particularly opportune time.”
Tim Costa, General Manager of Industrial Engineering and Quantum, NVIDIA, said, “The next era of scientific discovery will be shaped by systems that unite quantum computing and accelerated supercomputing, opening new paths to problems beyond the reach of either technology alone, with ROQUO, RIKEN is bringing that vision into operation by connecting NVIDIA GB200 NVL4 systems and NVIDIA Quantum-X800 InfiniBand with Fugaku and quantum computers through its Quantum-HPC Hybrid Platform.”
During pre-launch testing, ROQUO achieved 19.80 petaflops of double-precision (FP64) performance in the High Performance LINPACK benchmark. The system consists of 135 compute nodes equipped with NVIDIA GB200 NVL4 processors, providing a total of 540 NVIDIA Blackwell GPUs. The nodes are connected through NVIDIA Quantum-X800 InfiniBand networking, enabling communication speeds of up to 3.2 terabits per second and ROQUO also uses warm-water cooling technology intended to improve energy efficiency.