Technology Integration
Important
Medium
80% Confidence
NVIDIA and Emerald AI Demonstrate Dynamic Energy Adjustment in AI Factories
Summary
NVIDIA partners with Emerald AI to demonstrate grid-responsive energy management on a 96 Blackwell Ultra GPU cluster, using NVIDIA System Management Interface for real-time power telemetry and Emerald AI Conductor to dynamically adjust energy use while maintaining high-priority AI workload performance.
Key Takeaways
Testing at Nebius AI factory in London used 96 NVIDIA Blackwell Ultra GPUs connected via Quantum-X800 InfiniBand running production AI workloads. NVIDIA System Management Interface provided second-level GPU power telemetry.
During grid stress scenarios simulated by EPRI and UK National Grid (including lightning strikes and Euro Cup TV peak), the AI factory received signals via Emerald AI Conductor to temporarily reduce power consumption.
Tests showed 100% compliance on over 200 grid power targets while maintaining peak throughput for high-priority AI workloads and temporarily slowing flexible tasks.
During grid stress scenarios simulated by EPRI and UK National Grid (including lightning strikes and Euro Cup TV peak), the AI factory received signals via Emerald AI Conductor to temporarily reduce power consumption.
Tests showed 100% compliance on over 200 grid power targets while maintaining peak throughput for high-priority AI workloads and temporarily slowing flexible tasks.
Why It Matters
NVIDIA demonstrates AI infrastructure-energy system coordination capability, providing new paradigm for energy efficiency management in large-scale AI deployments that may influence future AI factory architecture....