Vertiv, a global provider of critical digital infrastructure and services, has introduced a new energy-efficient cooling and power reference architecture designed for Nvidia’s GB300 NVL72 platform. The announcement is a key step in enabling data centers to meet the growing demands of artificial intelligence (AI) workloads.
The reference architecture supports up to 142kW per rack and includes an end-to-end design for both power and cooling. According to Vertiv, this shift moves away from traditional data center buildouts, offering a more agile and efficient approach.
“Building upon our strong collaboration with Nvidia and a shared vision to advance AI infrastructure, Vertiv is proud to release our comprehensive reference design and SimReady 3D assets for the Nvidia GB300 NVL72 platform,” said Scott Armul, executive vice president of Global Portfolio and Business Units at Vertiv.
The architecture is designed to support higher-density data center needs, including power infrastructure capable of handling 800 VDC for 1 MW IT racks. Vertiv plans to roll out these solutions starting in 2026.
Among the key features of Vertiv’s architecture is the integration of Nvidia Omniverse technologies, allowing real-time collaboration among IT, facilities, and engineering teams. This helps customers simulate and optimize system performance before deployment.
The design also supports faster builds, with up to 1.5 times more AI performance, 50% quicker site deployment, and a 30% reduction in physical space. It is compatible with liquid and hybrid cooling systems and can achieve up to 70 percent better annual energy efficiency through higher water temperatures.
With a global team of more than 4,000 field service engineers, Vertiv said the architecture offers scalability and reliability for large-scale AI data centers worldwide.
“By combining Nvidia’s advanced AI platforms with Vertiv’s expertise in power and cooling infrastructure, we’re enabling customers to deploy next-generation data centers that are more efficient, scalable, and ready for the most demanding AI workloads,” said Dion Harris, senior director of HPC and AI Infrastructure at Nvidia.
Get the latest before it trends. Follow Back End News on LinkedIn, Facebook, X, YouTube, and TikTok for updates and in-depth coverage across the tech and security landscape.