AVEVA Collaborates with NVIDIA to Develop New Lifecycle Digital Twin Architecture for Gigawatt-Scale AI Factories

・Collaboration with NVIDIA accelerates GPU efficiency and the rapid, large-scale deployment of AI factories. ・AVEVA's engineering and operations software is integrated into the NVIDIA Omniverse DSX Blueprint to advance digital twins for AI factories. AVEVA, a global leader in industrial software driving digital transformation and sustainability, announced a new collaboration with NVIDIA on March 16 (UK time). The two companies are jointly building physical and digital modules deployable in large-scale data centers, known as AI factories, by integrating AVEVA's engineering and operations software into the NVIDIA Omniverse DSX Blueprint. This approach follows methodologies used in Engineering, Procurement, and Construction (EPC) projects. This initiative leverages AVEVA's comprehensive portfolio, particularly the CONNECT industrial intelligence platform and industrial digital twin technology. By utilizing domain-specific simulations, digital visualization, and collaborative design tools, the partnership aims to reduce the time-to-token for AI factories and enable high-speed, large-scale deployment while maximizing GPU utilization efficiency. The collaboration is being advanced in coordination with strategic partners Schneider Electric and ETAP, comprehensively promoting the design, simulation, construction, operation, and optimization of AI factories using the NVIDIA Omniverse DSX Blueprint. This extends the expertise and methodologies AVEVA has cultivated across many industrial sectors into the AI infrastructure domain. Delivering Digital Twin Value Across the AI Factory Lifecycle By incorporating its solutions into the Omniverse DSX Blueprint, AVEVA provides the following value throughout the AI factory lifecycle: ・New converters added to AVEVA Unified Engineering, enabling the ingestion of OpenUSD-based SimReady assets. This allows for the reuse of existing assets, new designs, and the utilization of high-precision data via the Omniverse library. ・AVEVA Asset Information Management provides a "Single Source of Truth," enabling seamless change management of facility and system information from design through operation. ・AVEVA Process Simulation enables the modeling and simulation of advanced liquid cooling networks for AI factories, supporting design optimization and maximizing cooling efficiency. ・AVEVA PI System enables the aggregation of IT and OT data across the NVIDIA Omniverse DSX Exchange. In the future, it will integrate with NVIDIA's NV-Tesseract model to analyze telemetry from BMS, EPMS, cooling systems, server racks, and workloads in real-time at scale, significantly transforming efficiency across the entire lifecycle. ・AVEVA Operations Control / Unified Operations Center provides a single, scalable platform for data center operations, integrating electrical (UPS, switchgear, PDU, generators, etc.), mechanical (chillers, CDU, etc.), and safety systems.