NVIDIA Opens Vera Rubin Rack Designs to OCP for AI Factory Infrastructure
At the Open Compute Project (OCP) Global Summit on October 12, NVIDIA announced that it is contributing Vera Rubin rack and compute tray designs to OCP through its open NVIDIA MGX architecture, enabling system makers to develop infrastructure for next-generation AI factories.
The updated MGX 1.2 specification adds rack-level blind-mate liquid cooling, interoperable rack interconnects, liquid-cooled 5000A busbars, and native 19-inch device support, allowing deployments of up to 250kW per rack while simplifying assembly, testing, maintenance, and deployment.
NVIDIA also released four open-source software components for DSX AI Factory, covering switch configuration, GPU cluster stress testing, rack firmware management, and bare-metal deployment. These components, derived from NVIDIA DGX Cloud operations, are designed to help cloud service providers turn hardware into production clusters within hours.
In collaboration with AMD, Google, and Meta, NVIDIA published Open Data Center Specification v0.7 to unify power, cooling, and physical layout designs for AI factories. Separately, NVIDIA worked with Google, Meta, and Microsoft on a grid resilience specification requiring large-scale AI facilities to stay connected during grid disturbances and recover to near full capacity in under two seconds. Meta also plans to adopt NVIDIA Spectrum-6 102.4Tb/s Ethernet switches to build an open-network Vera Rubin AI factory, with support for clusters of more than 100,000 GPUs.
