International.Vertiv will collaborate with NVIDIA and iGenius, to deploy Colosseum, one of the world's largest NVIDIA DGX AI supercomputers with NVIDIA Grace Blackwell superchips.
With a planned implementation in 2025 in Italy, Colosseum will transform the digital landscape as the first sovereign AI data center of its kind for regulated workloads.
Colosseum has been designed to meet the needs of highly regulated industries, such as financial services, healthcare, and public administration, and will comprise a fusion of transformative computing power, energy efficiency, and data sovereignty, while offsetting stringent data security requirements.
Colosseum, an NVIDIA DGX SuperPOD™, is the latest development in a long-standing collaboration between Vertiv and NVIDIA, and is strategically located in southern Italy to meet the needs of regional government, marking an important milestone in the AI landscape in Europe.
A modular design, for efficiency
Colosseum combines Vertiv's expertise in infrastructure management, accelerated computing, and NVIDIA's Omniverse Blueprint for AI factory design and operations. The deployment will leverage Vertiv's 360AI reference architecture infrastructure platform for data center power and cooling designed for the NVIDIA GB200 NVL72, co-developed with NVIDIA and introduced in late 2024.
This modular and scalable system positions iGenius to deploy one of the fastest and largest hyperscale AI supercomputers to support sovereign AI.
In addition, Vertiv has expanded its library of reference designs in its AI Hub with the jointly developed power and cooling design for the NVIDIA GB300 NVL72. Staying a generation ahead in GPU technology, Vertiv enables customers to plan infrastructure ahead of the advent of silicon, with deployment-ready designs ready for higher rack power densities, as well as repeatable models for AI factories at scale.
Simulation with NVIDIA Omniverse. A fast delivery.
Colosseum has been co-designed as a physically-accurate digital twin developed with NVIDIA Omniverse technologies; this has facilitated real-time collaboration between Vertiv, iGenius, and NVIDIA, to streamline system-level decisions and reduce the cycle from design to deployment.
The Omniverse Blueprint enables real-time simulations, and engineers can test and refine designs immediately, rather than waiting for lengthy simulation processes, reducing simulation times from months to hours. In addition, Vertiv's factory integration and manufacturing processes reduce implementation times by up to 50%, compared to traditional data center builds.
This 3D design collaboration validated the entire infrastructure set, enabled predictive modeling of thermal load, electrical flow, and site layout—from 132 kW liquid-cooled racks to modular power systems—before building a single module.
Intelligently designed. Unified by software.
Vertiv designs, manufactures, delivers, installs, and commissions this AI-ready, prefabricated modular data center solution. It includes power, cooling, management, monitoring, servicing, and maintenance, with an initial power and cooling capacity of up to 132 kW per rack and scalable for future designs.
The structure integrates prefabricated free space inside and modular gray space outside, to deliver exceptional scalability and energy efficiency, as well as transform the way data centers are built and deployed.
Colosseum will leverage NVIDIA Mission Control solutions for data center operations and Vertiv™ Unify to simplify and synchronize building management for AI factories. Vertiv Unify enables:
● Real-time coordination of power, cooling and computing
● Synchronization with digital twins for closed-loop optimization
● AI-ready capabilities that enable independent decision-making
By integrating NVIDIA Omniverse technologies, Vertiv Unify enables real-time updates between physical systems and digital models, facilitating predictive maintenance, what-if simulations, and scenario testing before operational risks occur.
The plan for AI factories around the world
Colosseum is more than just a data center; is the blueprint for scalable and repeatable sovereign AI factories. By combining cloud-scale density, local data control, and modular deployment, it lays the groundwork for the next phase of AI, where inference will need to be secure, fast, compliant, and distributed.
It is not a single project, but a point of reference. iGenius is creating a plan with Colosseum designed to be replicated around the world, hand-in-hand with Vertiv and NVIDIA for future platforms, along with the DGX GB300 systems and beyond. The future of sovereign AI is no longer a theoretical concept; it's being built right now.
