Jacobs has secured a three-year software-as-a-service contract with Nvidia to implement its Data Center Digital Twin platform at a large-scale AI research and development (R&D) site in the US.
The deployment is designed to assist facility operators with performance monitoring and operational planning across the research facility.
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
The digital twin system, which uses Nvidia Omniverse libraries, connects engineering models, operational technology, and real-time site data into a single operational interface.
This arrangement enables data centre teams to run predictive simulations, track infrastructure conditions, and test operational scenarios.
Supported applications include energy forecasting, dynamic power load balancing, predictive maintenance, liquid coolant leak detection, and staff training.
The system also allows operators to model changing AI workloads to assess cooling requirements and power availability.
Jacobs executive vice-president Amer Battikhi said: “As digital infrastructure rapidly scales, operators need smarter and more connected ways to optimise performance and manage energy demand, while improving resiliency.
“Jacobs is integrating real-time data and simulation technologies to help Nvidia advance the next generation of AI infrastructure management.
“The project reflects the growing role of digital twins in helping operators manage critical infrastructure environments.”
Jacobs is developing the digital twin platform into an operating system for AI facilities, using AI agents that analyse operational facility data to forecast conditions, propose actions, and manage operations with increasing autonomy.
The agreement builds on existing work between the two companies involving digital twin technology for computing infrastructure.
Jacobs is also delivering high-performance computing environments for other clients, including Hut 8 in the US, Start Campus’ SINES DC Campus in Portugal, and PsiQuantum in Australia.
Last month, Nvidia announced a collaboration with several companies in the Australian data centre sector to expand infrastructure supporting AI workloads, targeting up to 2GW of new capacity by 2027.
