NVIDIA® H100 InfiniBand Networking Solution Based on PicOS®

PHILISUN A next-generation network architecture specifically designed for high-performance computing (HPC) and artificial intelligence (AI), combining the powerful computing capabilities of NVIDIA H100 GPUs with the ultra-low latency communication of InfiniBand, enabling intelligent and automated network management through PicOS® software and the AmpCon-DC management platform.

High-Performance Computing Network

Based on the NVIDIA® H100 GPU, the H100 InfiniBand solution utilizes PicOS® software and the AmpCon-DC management platform, tailored to the network topology of HPC architecture—including InfiniBand networks, management networks, and storage networks—to accommodate various business needs.

InfiniBand Network Management Network Storage Network

InfiniBand Network

The ultimate network crafted with NVIDIA® H100 GPUs and InfiniBand switches not only delivers ultra-low latency and high bandwidth but also ensures flawless data transmission during high-speed exchanges through advanced lossless technology.

Management Network

Through the PicOS® and AmpCon-DC management platforms, full lifecycle automated management—from configuration and monitoring to preventive fault exclusion—is achieved, effectively enhancing infrastructure utilization and reducing operational costs.

Storage Network

The built-in BGP protocol delivers exceptional routing control, ensuring low-latency, highly available forwarding paths for critical services such as storage. Its flexible scalability further enables it to meet your specific requirements for capacity and bandwidth at any time.