Pilot at Portugal’s Sines data centre campus combines digital twin technology and AI infrastructure modelling to accelerate deployment, improve operational efficiency and optimise token production.
Schneider Electric, AVEVA, Nscale and Start Campus have announced an end-to-end digital twin pilot designed to accelerate the deployment and optimisation of large-scale AI data centres.
Built using NVIDIA’s Omniverse DSX Blueprint, the Immersive Twin integrates technologies from Schneider Electric, AVEVA and ETAP to connect AI workloads with power, cooling, electrical and operational models. The platform enables operators to simulate infrastructure configurations and assess potential changes before implementing them on-site.
The pilot focuses on Nscale’s operations at the SIN01 data centre within Start Campus’ 1.2 GW Sines Data Campus in Portugal. The campus is designed to accommodate successive generations of AI infrastructure and is expected to host more than 66,000 NVIDIA Rubin GPUs.
By combining engineering data, operational information and workload requirements, the digital twin aims to identify infrastructure bottlenecks, evaluate upgrades and improve token efficiency while reducing deployment risks.
“By bringing together power, cooling and infrastructure in one environment, operators can simulate upgrades, expansions and greenfield projects while continuously optimizing physical operations as conditions change, improving both ‘time to token’ and overall efficiency.” — Caspar Herzberg, CEO, AVEVA
“By bringing together power, cooling and infrastructure in one environment, operators can simulate upgrades, expansions and greenfield projects while continuously optimizing physical operations as conditions change,” said Caspar Herzberg, CEO of AVEVA. He added that the platform provides simulation and real-time asset performance visibility through technologies including CONNECT, PI System, AVEVA Process Simulation, AVEVA Unified Engineering and AVEVA Asset Information Management.
The OpenUSD-based platform supports what-if scenario testing across power distribution, cooling and operational processes. It is designed to help teams validate configurations, assess upgrade impacts and coordinate infrastructure decisions before construction or modification begins.
For Nscale, the approach is expected to improve time to token, operational flexibility and infrastructure efficiency. Start Campus provides the real-world environment for demonstrating how digital twins can help coordinate compute, power and cooling as a unified system.
The initiative reflects a broader industry shift towards digitally integrated AI factories, where infrastructure is designed and optimised holistically to support growing AI workloads.
