Vertiv Brings Converged Bodily Infrastructure to NVIDIA Vera Rubin DSX AI Factories

Vertiv (NYSE: VRT), a worldwide chief in important digital infrastructure, at the moment introduced its position in advancing converged bodily infrastructure designs for the NVIDIA Vera Rubin DSX AI manufacturing facility reference design and the NVIDIA Omniverse DSX Blueprint.

As AI factories scale in density, complexity, and energy demand, operators are beneath strain to compress time to deployment, enhance infrastructure utilization, and cut back integration danger. A brand new infrastructure design method that reduces complexity, improves confidence earlier than buildout, and accelerates time to capability is now out there to satisfy these evolving wants. By way of its work with NVIDIA, Vertiv is contributing simulation-ready, or DSX SimReady digital energy and cooling property, validated interfaces, and repeatable infrastructure constructing blocks designed to assist clients deploy AI factories sooner and with better operational assurance.

This work displays an enlargement of Vertiv’s established method to converged bodily infrastructure—a system-level mannequin that integrates energy, cooling, controls, and providers into interdependent designs optimized throughout the total energy practice and thermal chain. This method is enabled by means of 5 foundational components: repeatable constructing blocks, outlined interfaces, system orchestration, digital continuity, and lifecycle help. Collectively, these components help extra scalable AI manufacturing facility execution by serving to cut back design complexity, strengthen coordination throughout infrastructure domains, and enhance confidence from preliminary design by means of deployment and operation.

On the core of this method is a scalable constructing block structure designed across the standardized 12.5MW infrastructure blocks of Vertiv™ OneCore built-in modular options that may be mixed, configured, and prolonged to help deployments starting from smaller AI clusters to gigawatt-scale AI factories. By establishing repeatable block-level designs with validated interfaces, Vertiv goals to simplify scaling whereas enhancing deployment consistency, system coordination, and operational efficiency.

“AI factories are forcing a basic change in how digital infrastructure is designed, validated, and deployed,” stated Scott Armul, chief product and know-how officer at Vertiv. “Vertiv’s position is to assist flip complicated AI infrastructure from a set of separate merchandise into converged, simulation-ready bodily programs. Working with NVIDIA, we’re serving to clients transfer sooner from design to deployment. By combining our energy and cooling portfolio with validated interfaces and digital fashions, we will help clients speed up growth, enhance operational confidence, and unlock higher output per watt.”

Vertiv’s collaboration helps the event of digitally validated AI manufacturing facility infrastructure utilizing real-time simulation and system-level modeling earlier than bodily deployment begins. This method is designed to assist clients:

• cut back deployment complexity and subject integration danger,
• speed up time to operational readiness,
• enhance infrastructure coordination throughout energy, cooling, and controls,
• and optimize efficiency from grid connection by means of chip-level thermal administration and heat-reuse pathways.

Vertiv’s contribution is grounded in its means to carry collectively one of many business’s most full portfolios of important energy, thermal administration, built-in controls, and lifecycle providers right into a cohesive converged bodily infrastructure. In contrast to standard modular or prefabricated approaches that primarily compress schedule, converged bodily infrastructure is meant to ship each deployment velocity and compounding system-level positive factors. By standardizing interfaces and creating repeatable constructing blocks, Vertiv goals to help extra scalable AI manufacturing facility execution whereas enabling improved efficiency, effectivity, and reliability.

“As AI factories scale to unprecedented ranges of energy and density, enterprises require a converged method to bodily infrastructure that unifies energy, cooling, and digital twin simulation to scale back deployment danger,” stated Vladimir Troy, vp of AI Infrastructure at NVIDIA. “By integrating simulation-ready infrastructure fashions into the NVIDIA Vera Rubin DSX design, Vertiv is offering the repeatable constructing blocks and validated interfaces essential to speed up the trail from design to operational readiness.”

This collaborative output, Vertiv™ OneCore Rubin DSX, is a design consequence grounded in converged bodily infrastructure that Vertiv will proceed to iterate for a number of compute generations forward. It’s meant to help AI manufacturing facility builders with parameterized infrastructure fashions and deployment-ready constructing blocks that span energy, cooling, controls, and lifecycle providers.

Vertiv expects this work to tell future converged infrastructure choices throughout hyperscale, colocation, enterprise, and rising AI deployment environments.

For extra info on Vertiv options for AI infrastructure, go to Vertiv.com.