ASUS Empowers Taiwan’s NCHC with Nano4 NVIDIA HGX H200 AI Supercomputer Boosting AI Capabilities
ASUS has equipped Taiwan’s National Center for High-performance Computing with the Nano4 NVIDIA HGX H200 AI supercomputer delivering up to 81.55 PFLOPS, ranked No. 29 on the TOP500, enhancing Taiwan’s AI and scientific computing power.
ASUS has propelled Taiwan’s National Center for High-performance Computing (NCHC) into a new era of AI-powered research with the deployment of the Nano4 NVIDIA HGX H200 AI supercomputer system. This cutting-edge infrastructure offers up to 81.55 petaflops (PFLOPS) of processing power, placing it at No. 29 on the prestigious TOP500 list of the world’s fastest supercomputers, a significant leap for Taiwan’s AI and scientific computing capabilities.
The Nano4 system features ASUS-designed ESC NM2N721-E1 and ESC N8-E11V servers integrating the latest NVIDIA HGX H200 architecture. Each rack combines 36 NVIDIA Grace CPUs with 72 NVIDIA Blackwell GPUs, interconnected via the NVIDIA NVLink technology, enabling high-bandwidth and low-latency data transfer critical for large-scale AI workloads. This architecture supports diverse AI applications including genomics, climate modeling, quantum information, and industrial automation.
Two key elements bolster the system’s performance: the powerful NVIDIA HGX H200 GPUs equipped with 141GB of HBM3e memory delivering a bandwidth of 4.8 terabytes per second, and an advanced Direct-to-Chip (D2C) liquid cooling solution ensuring optimal thermal performance and energy efficiency even under demanding AI workloads.
This deployment strengthens Taiwan’s position in global AI research and development by providing researchers and enterprises with unprecedented computing resources, accelerating innovation across academia and industry fronts.
