Hongbai Technology, a subsidiary of IFF, attended the 2024 Open Computing Program Global Summit held at the San Jose Convention Center in California today to showcase its fully integrated AI product line to empower future AI data centers. The NVIDIA HGX B200 liquid-cooled AI accelerator presented by Hongbai provides powerful AI training performance, can be seamlessly and perfectly integrated into the OCP ORv3 rack and deployed in the customer's data center. Hongbai also showcased NVIDIA GB200 NVL72 and water-to-water CDU solutions. The GB200 NVL72 is a liquid-cooled cabinet specially designed for mega-parameter large language model training and real-time inference. It is equivalent to a giant GPU with 72 Blackwell Tensor core GPUs connected to NVLink.

Zhitongcaijing · 10/16 03:25
Hongbai Technology, a subsidiary of IFF, attended the 2024 Open Computing Program Global Summit held at the San Jose Convention Center in California today to showcase its fully integrated AI product line to empower future AI data centers. The NVIDIA HGX B200 liquid-cooled AI accelerator presented by Hongbai provides powerful AI training performance, can be seamlessly and perfectly integrated into the OCP ORv3 rack and deployed in the customer's data center. Hongbai also showcased NVIDIA GB200 NVL72 and water-to-water CDU solutions. The GB200 NVL72 is a liquid-cooled cabinet specially designed for mega-parameter large language model training and real-time inference. It is equivalent to a giant GPU with 72 Blackwell Tensor core GPUs connected to NVLink.