As GPU demands for AI and HPC applications increase, Supermicro has introduced a new datacenter-scale liquid cooling solution designed to meet the needs of the upcoming Nvidia Blackwell servers. This solution is a comprehensive package that includes advanced cooling components, software, and management tools to optimize performance and sustainability.

The key feature of Supermicro's liquid cooling solution is its ability to support up to 96 B200 GPUs per rack, which is a significant increase from traditional setups. This solution is tailored for ultra-dense AI servers, capable of housing dual top-tier CPUs and up to eight Nvidia HGX GPUs in a 4U configuration.

One of the standout benefits of this solution is its rapid deployment capability, which is crucial for companies looking to upgrade or build new facilities quickly. The technology is engineered to handle extreme heat levels, efficiently cooling servers that demand up to 12kW of power and AI racks generating over 100kW of heat.

Included in the package is the SuperCloud Composer software, which provides real-time monitoring and control over all components. This software helps data center operators manage costs and maintain system reliability, ensuring smooth and efficient operations.

Supermicro has also implemented this liquid cooling solution in some of the largest AI facilities ever built, showcasing its capability to meet the technical and financial goals of data center operators for both new construction and modernization of existing data centers.