Vertiv has introduced Vertiv SmartIT, a pre-engineered infrastructure platform designed for high-density AI deployments in existing facilities and new data centers. Available globally, the solution integrates power distribution, direct-to-chip liquid cooling, rack infrastructure and monitoring in a factory-validated architecture for enterprises, research institutions and colocation providers.
Previously deployed with NVIDIA to support the NVIDIA DGX GB300 system, SmartIT combines DC power conversion and distribution, in-rack cooling distribution and monitoring. Factory validation reduces the need for on-site integration, with commissioning focused on connections to facility power and cooling infrastructure. The platform can be deployed in existing enterprise facilities, commercial buildings, new data centers and hyperscale environments.
SmartIT supports NVIDIA GB300 NVL72, NVIDIA Vera Rubin NVL72, NVIDIA DGX Rubin NVL8 and NVIDIA HGX Rubin NVL8 systems based on NVIDIA MGX and Open Compute Project (OCP) ORV3 architectures. The platform scales from a single rack to multi-rack configurations without redesign between expansion phases. Integrated leak detection, automated response and continuous monitoring support operations, with compatibility for Redfish, SNMP and Modbus protocols.
The platform can also be combined with Vertiv SmartRun for row-level integration, Vertiv rack power distribution units for power distribution and monitoring, and Vertiv Unify for monitoring, analytics and controls across power and thermal infrastructure. Vertiv Global Services provides deployment, commissioning and lifecycle management support. At NVIDIA GTC Berlin in October 2026, Vertiv demonstrated SmartIT with an NVIDIA HGX Rubin NVL8 demo unit, showcasing an architecture designed to support NVIDIA DGX Rubin NVL8 deployments.
“Vertiv SmartIT reflects our approach to solving that challenge by bringing these disciplines together in a factory-validated architecture,” said Scott Armul, Chief Product and Technology Officer at Vertiv.
