Supermicro Expands OCP Portfolio to Accelerate AI Data Center Deployments

Supermicro has expanded its Open Compute Project (OCP)-based portfolio with new systems, racks, and cooling solutions aimed at improving the deployment of artificial intelligence and high-performance computing (HPC) infrastructure. 

Announced on October 13, 2026, the expanded lineup integrates computing, power, networking, and cooling components through the company’s Data Center Building Block Solutions (DCBBS).

Integrated Rack-Scale Infrastructure

The portfolio includes ORv3 and ORW racks featuring centralized 48V DC power distribution, alongside power shelves, battery backup units, networking switches, and system management software. Using these components, Supermicro aims to simplify data center integration and reduce deployment complexity.

The lineup also features rack-scale platforms for demanding AI workloads, including NVIDIA Vera Rubin NVL72 systems and liquid-cooled NVIDIA HGX B300 solutions designed for OCP-compliant racks.

Liquid Cooling

To address the thermal demands of modern AI infrastructure, Supermicro offers direct liquid cooling and rear door heat exchangers (RDHx), with cooling capacities ranging from 10 kW to 120 kW. 

These solutions are designed to help operators manage heat in high-density environments, including facilities adapting existing infrastructure for AI workloads.

Faster Data Center Deployment

By integrating servers, power delivery, cooling, and management into coordinated rack-scale solutions, Supermicro aims to shorten deployment timelines and improve power and cooling efficiency. 

The company is presenting its expanded portfolio at the OCP Global Summit in San Jose, California, taking place October 12 – 15, 2026.