The rise of artificial intelligence is reshaping data centre infrastructure at an unprecedented pace. With AI workloads demanding ever greater computational power, the challenge of managing energy consumption has become a critical issue. In response, Schneider Electric and ETAP have unveiled a new digital twin technology designed to provide real-time simulation and analysis of power requirements, extending from the electricity grid to individual AI chips.
Built using NVIDIA Omniverse, the collaboration introduces a new level of precision in modelling and optimising data centre power infrastructure. By integrating real-time power system data with advanced analytics, the digital twin enables data centre operators to simulate energy demand across mechanical, thermal, networking, and electrical systems. This comprehensive approach offers the potential to significantly improve efficiency, reduce costs, and enhance reliability.
Enhancing power system precision
Unlike previous solutions, which provided only limited visibility into electrical systems, ETAP’s digital twin creates a fully interactive virtual replica of a data centre’s power infrastructure. This allows operators to analyse real-time power consumption patterns, assess infrastructure needs, and conduct dynamic ‘what-if’ scenario analysis. The ability to predict and manage energy use at this level of detail is increasingly essential as AI models become more complex and power intensive.
AI workloads, particularly those associated with model training and high-performance inference tasks, are pushing data centres towards higher rack power densities. Traditional power estimation methods provide only average consumption figures at the rack level, but ETAP’s digital twin aims to enhance precision by modelling dynamic load behaviour at the chip level. This enables more effective power system design and optimisation, helping data centres balance performance with energy efficiency.
Addressing the energy challenge of AI
The ‘grid to chip’ approach pioneered through this collaboration is expected to have wide-reaching implications. By bridging electrical engineering with advanced AI-powered simulation, the technology addresses key challenges associated with AI adoption, including power distribution, infrastructure reliability, and long-term sustainability.
“As AI workloads grow in complexity and scale, precise power management is critical to ensuring efficiency, reliability, and sustainability,” Dion Harris, senior director of HPC and AI factory solutions at NVIDIA, said. “Through our collaboration with ETAP and Schneider Electric, we’re offering data centre operators unprecedented visibility and control over power dynamics, empowering them to optimise their infrastructure and accelerate AI adoption while enhancing operational resilience.”
The ability to optimise data centre operations in real time will become increasingly important as AI adoption accelerates. Start-ups, enterprises, and colocation providers are already facing rising energy costs and growing regulatory pressure to improve sustainability. Digital twin technology provides a means to not only reduce total cost of ownership but also improve grid reliability and resilience in an era of rising demand.
Tanuj Khandelwal, CEO of ETAP, believes this collaboration is about more than just technological innovation. “We’re fundamentally reimagining how data centres can be designed, managed, and optimised in the AI era. By bridging electrical engineering with advanced virtualisation and AI technologies, we’re creating a new paradigm for infrastructure management.”
Pankaj Sharma, executive vice president for data centres, networks and services at Schneider Electric, emphasised the broader industry implications of this initiative. “Collaboration, speed, and innovation are the driving forces behind the digital infrastructure transformation that’s required to accommodate AI workloads,” he said. “Together, ETAP, Schneider Electric, and NVIDIA are not just advancing data centre technology , we are empowering businesses to optimise operations and seamlessly navigate the power requirements of AI.”
As AI continues to reshape industries, the need for intelligent infrastructure capable of handling the associated energy demands will only grow. With digital twin technology, data centre operators now have the tools to design and manage power systems with greater precision, ensuring they are prepared for the next wave of AI-driven transformation.
