Vertiv, a global leader in critical digital infrastructure, announced Tuesday the launch of its gigawatt-scale reference architectures tailored for NVIDIA’s Omniverse DSX Blueprint, a move designed to accelerate generative AI deployments and reduce Time to First Token on platforms like NVIDIA Vera Rubin.
The announcement coincides with NVIDIA’s unveiling of its AI Factory Research Center in Manassas, Virginia, positioning Vertiv as a key enabler in building multi-generational, gigawatt-class AI factories that address surging demands for power, cooling, and scalability in the AI industrial revolution.
Unlike traditional one-size-fits-all designs, Vertiv’s architectures offer flexible deployment options—including stick-built, hybrid, and fully prefabricated models — via the pre-validated Vertiv OneCore platform, to suit the varied needs of customers for speed and efficiency.
Key innovations include optimized “Grid-to-Chip” power topologies for smaller footprints and higher resilience, advanced liquid cooling systems with chip-to-heat reuse for extreme thermal loads, and SimReady 3D assets for digital twin simulations that streamline pre-construction planning.
ALSO READ: FORTINET MARKS 25TH ANNIVERSARY IN CYBERSECURITY
The prefabricated OneCore approach integrates compute, power, cooling, and services holistically, potentially compressing delivery timelines by 50% compared to conventional methods while enhancing space utilization.
“The AI industrial revolution requires a complete reinvention of data center infrastructure,” said Scott Armul, executive vice president of global portfolio and business units at Vertiv.
“Vertiv’s collaboration with NVIDIA is foundational to our mission of equipping our customers with the most advanced and flexible solutions to navigate this transformation.”
NVIDIA’s Dion Harris, senior director of HPC, Cloud, and Infrastructure Solutions, echoed the partnership’s significance: “NVIDIA and Vertiv are collaborating to develop flexible, rapidly-deployable reference architectures that deliver the critical power and cooling foundation needed to scale the next generation of gigawatt-class AI factories.”
Backed by Vertiv’s extensive power and cooling portfolio and a global network of over 4,000 field engineers, these architectures are already in use for major AI factory projects, promising energy-efficient, high-availability operations amid the global AI boom.
