As NVIDIA unveils its AI Factory Research Center in Virginia and the Omniverse DSX Blueprint for gigawatt-scale, multi-generational AI infrastructure, Vertiv is playing a pivotal role in enabling the next wave of AI innovation.

Vertiv’s reference architectures for the NVIDIA Omniverse DSX Blueprint offer a wide range of deployment options, from traditional models and hybrid solutions to fully prefabricated configurations, to meet diverse customer needs for speed, flexibility, and scale. To further accelerate the design process, Vertiv facilitates the integration of virtual and physical systems through its SimReady 3D asset library, enabling the creation of digital twins of AI factories before construction begins.

The prefabricated version of the architecture is built on the prevalidated Vertiv™ OneCore platform, which envisions the entire facility as a single, co-engineered system where compute, power, cooling, and services are fully integrated. The Vertiv OneCore platform delivers key benefits for gigawatt-scale deployments, such as reducing lead times by up to 50% compared to traditional construction methods, while also optimizing space and performance.

“The AI ​​industrial revolution requires a complete reinvention of data center infrastructure. Vertiv’s collaboration with NVIDIA is fundamental to our mission to provide our customers with the most advanced and flexible solutions to address this transformation,” said Scott Armul, executive vice president of Vertiv’s global portfolio and business units. “Vertiv’s reference architecture for NVIDIA Omniverse DSX is more than just a blueprint: it’s a comprehensive approach that accelerates AI deployment to a previously unimaginable scale. With our complete grid-to-chip solutions, flexible deployment models like Vertiv™ OneCore, and the power of digital twins within the NVIDIA Omniverse libraries, we are enabling the factories of the future, today.”

Vertiv's reference architectures incorporate several key innovations:

Optimized power topologies: Vertiv's innovative grid-to-chip power system is designed to enhance the NVIDIA Omniverse DSX Blueprint, delivering a more efficient system with a smaller physical footprint, capable of providing the resilience and scalability required by multi-generational AI platforms.
Advanced liquid cooling solutions: The architecture integrates Vertiv's industry-leading liquid cooling systems, designed to handle the high thermal demands of accelerated computing. This advanced thermal solution, combined with a full chip-to-heat reuse thermal chain, enables a design perfectly tailored for maximum performance across current and future processor generations.
Deployment flexibility: Recognizing that every AI implementation is unique, Vertiv's reference architecture is the first to offer a validated design for traditional, hybrid, and prefabricated build methodologies. This flexibility can reduce Time to First Token by up to 50% compared to conventional build methods, allowing organizations to choose the most appropriate path based on their specific objectives.

“An efficient, high-performance AI infrastructure is essential for the next industrial revolution, and its development requires a robust partner ecosystem,” said Dion Harris, senior director of HPC, Cloud, and Infrastructure at NVIDIA. “NVIDIA and Vertiv are collaborating to develop flexible, rapidly deployable reference architectures that provide the critical power and cooling foundation needed to scale the next generation of AI factories to gigawatt scale.”

This comprehensive range of architectures is backed by the industry’s most complete power and cooling portfolio and complemented by a global network of more than 4,000 field technicians, providing the foundation for deploying and operating highly efficient, available, and scalable AI factories. Vertiv is currently involved in the design phases of several large-scale AI factory projects using the Vertiv™ OneCore prefabricated design, tailored to meet gigawatt-scale demands.

More information