< Back to all clusters
[TECHNOLOGY] · Taiwan · 21 sources

started · updated

Changhua Christian Hospital launches NVIDIA B200-powered AI platform

Changhua Christian Hospital (CCH) has officially launched a high-performance computing platform powered by NVIDIA B200 GPUs to enhance its artificial intelligence capabilities. The new infrastructure utilizes Dell PowerEdge XE9780 servers equipped with NVIDIA HGX B200 8-GPU systems and NVLink high-speed interconnect technology.

This upgrade is designed to support diverse AI workloads, including large language models, predictive AI, computer vision, and multimodal AI. The platform is deployed on-site to ensure data security and privacy in compliance with hospital governance standards. CCH aims to integrate these tools into existing medical information systems to assist clinical care, information processing, and team collaboration.

Looking forward, the hospital plans to develop Agentic AI, which will allow AI to plan steps, select tools, and connect with hospital systems under human supervision to coordinate multi-step tasks. The ultimate goal is to reduce repetitive workloads for medical staff, allowing them to focus more on patient care.

Entities

Changhua Christian Hospital · Chen Mu-kuan · Cornelis Networks · Dell · Dell Technologies · Fixstars · GetWorks · HPE · NTTPC Communications · Nvidia · Qualcomm

Claims

What the coverage asserts, and how many sources carry each claim.

  • [○ 1 SOURCE] The system includes NVIDIA HGX B300 GPUs within HPE ProLiant Compute XD685 servers.
  • [○ 1 SOURCE] Three companies launched the 'Water-cooled GPU Container DC-PKG' on September 8, 2026.
  • [○ 1 SOURCE] The on-premise AI infrastructure starts at an initial cost of 300 million yen.
  • [○ 1 SOURCE] The monthly fee for the small configuration starts from 750,000 yen.
  • [○ 1 SOURCE] The shortest delivery time for the containerized data center is eight months.

Sources

智能体把压力给到通用计算:Arm的答案是可定制,也可开箱即用_天极网 [yesky.com]
1 day ago
about 20 hours ago