Schneider, Aveva and Nvidia Develop Blueprints for AI Factories

Key Takeaways

• Schneider, Aveva, Nvidia co‑develop Vera Rubin AI data‑center design.
• Design validates power, cooling, and digital‑twin integration for GPU racks.
• Higher 480 VAC and 45 °C loops boost efficiency and token‑per‑watt.
• Nemotron AI automates alarm management, cutting response times.
• Digital‑twin workflow reduces engineering cycles and accelerates deployment.

Pulse Analysis

The surge in generative‑AI workloads is forcing data‑center operators to rethink traditional infrastructure. Schneider Electric’s partnership with Aveva and Nvidia delivers a reference architecture that tackles the power‑density challenge by raising supply voltage to 480 VAC and allowing higher coolant temperatures. By aligning hardware specifications with Nvidia’s MaxQ and MaxP modes, the design maximizes tokens per watt, a critical metric for gigawatt‑scale AI factories where energy costs dominate total ownership.

Beyond hardware, the trio’s integration of Nvidia’s Omniverse DSX digital‑twin platform adds a simulation‑first workflow. Engineers can model electrical distribution, thermal dynamics and airflow in a virtual environment before any physical rack is built. This reduces the engineering cycle, accelerates time‑to‑token, and lowers risk by catching design flaws early. Aveva’s multi‑domain simulation tools further refine performance predictions, delivering a validated, performance‑optimized blueprint that can be reused across multiple sites.

The operational layer gains a boost from Nvidia’s Nemotron agentic AI, which autonomously interprets alarm streams and recommends corrective actions. Early tests show faster issue resolution and fewer unnecessary dispatches, translating to higher uptime for AI workloads. As AI models grow larger and more complex, such intelligent monitoring becomes essential for maintaining service levels while keeping energy consumption in check. The collaboration therefore sets a precedent for end‑to‑end, AI‑ready data‑center design that blends power efficiency, digital twins and autonomous operations.