SAN JOSE, CA, Sep 25, 2026 – Super Micro Computer is shipping NVIDIA Vera Rubin NVL72 racks integrated with its DLC-2 direct liquid cooling system. The racks use Supermicro’s Data Center Building Block Solutions (DCBBS), which covers compute, cooling, power and deployment services.
“We have spent years building the liquid-cooling stack, the manufacturing capacity, and the deployment teams for exactly this moment,” said Charles Liang, president and CEO, Supermicro. “Our customers can now order a Scalable Unit and receive production-ready systems with end-to-end integration, because we design and build every layer between the cold plate and the cooling tower. Our development of DLC-2 technology is what sets Supermicro apart and allows us to deliver the most complete solution for the NVIDIA Vera Rubin platform.”
NVIDIA designed Vera Rubin platform in conjunction with direct liquid cooling technology. The cooling loop transfers heat from cold plates to facility cooling towers. Supermicro tests and validates each rack with the liquid cooling system before deployment.
Supermicro handles mechanical engineering of individual compute nodes, rack manufacturing and cluster deployment using NVIDIA’s reference architecture. Its component portfolio includes cold plates, hose kits, rack power shelves and cooling distribution units (CDUs). Cooling equipment includes in-row and in-rack CDUs, liquid-to-air sidecar CDUs, rear-door heat exchangers and facility cooling towers.
Rack and Infrastructure Specifications
Supermicro’s DCBBS offering includes:
- NVIDIA Vera Rubin NVL72 racks: Each liquid-cooled rack combines 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs operating as one machine. Eighteen 1U compute trays each contain four Rubin GPUs and two Vera CPUs. Nine sixth-generation NVIDIA NVLink switch trays connect the compute trays and provide 216 TB/s of scale-up bandwidth. Each rack contains 20.7 TB of HBM4 memory and up to 54 TB of LPDDR5X memory.
- In-row cooling distribution units: Each Supermicro CDU is rated at 1.8 MW, deployed with N+1 redundancy.
- Optional rear-door heat exchangers: These capture residual heat from the racks.
- Networking and cabling services: Integration follows NVIDIA’s reference architecture and covers the AI compute fabric, converged fabric and management cabling.
Scalable Unit and Deployment Planning
The DCBBS Blueprint defines bill of materials for power requirements ranging from 5 MW to gigawatt scale.
A blueprint based on one Vera Rubin NVL72 unit includes 1,152 Rubin GPUs across 16 compute racks. The configuration provides 331 TB of HBM4 memory. It also specifies cooling capacity, power delivery, context memory storage and networking matched to the compute configuration.
Source: Supermicro
About Supermicro Computer
Super Micro Computer is a U.S. technology company founded in 1993 and headquartered in San Jose, CA. It designs and manufactures server and storage systems, networking devices and server management software. The company also develops components such as motherboards, power systems and chassis used in its hardware platforms. Supermicro provides products for enterprise data centers, cloud computing, AI, telecommunications and edge computing environments. Its portfolio includes servers, AI systems, storage platforms, IoT systems, switches and related support services. The company designs and manufactures products in facilities in the United States, Taiwan and the Netherlands. Its systems support a range of processors, memory configurations, GPUs, storage and networking options, along with air and liquid cooling methods. Supermicro serves global customers across enterprise and service provider markets. The company employs about 5,000 people worldwide.
