Empirically-calibrated H100 node power models for accurate AI training energy estimation
Open Access
- 8 December 2025
- journal article
- Published by IOP Publishing in Environmental Research: Energy
- Vol. 2 (4) , 045016
Abstract
Accurately quantifying the energy use of artificial intelligence (AI) training is critical for infrastructure planning, carbon accounting, and sustainable data center operation, but few studies have directly measured the power consumption of production workloads on contemporary hardware. By combining empirical measurements from Brookhaven National Laboratory during AI training on 8-graphics-processing-unit H100 systems with open-source benchmarking data, we develop statistical models relating computational intensity to node-level power consumption. We measure the gap between manufacturer-rated thermal design power (TDP) and actual power demand during AI training. Our analysis reveals that even computationally intensive workloads operate at only 76% of the 10.2 kW TDP rating. Our architecture-specific model, calibrated to floating-point operations, predicts energy consumption with 11.4% mean absolute percentage error, significantly outperforming TDP-based approaches (27%–37% error). We identified distinct power signatures between transformer and convolutional neural network architectures, with transformers showing characteristic fluctuations that may impact grid stability. These results provide a measurement-grounded basis for improving AI training energy estimates, enabling more reliable infrastructure sizing, cost projections, and environmental impact assessments.Keywords
All Related Versions
Funding Information
- Lawrence Berkeley National Laboratory (DE-AC02-05CH11231)
- U.S. Department of Energy for the Industrial Efficiency and Decarbonization Office
This publication has 65 references indexed in Scilit:
- Enhancing data center cooling efficiency and ability: A comprehensive review of direct liquid cooling technologiesEnergy, 2024
- To better understand AI’s growing energy use, analysts need a data revolutionJoule, 2024
- Estimating the environmental impact of Generative-AI services using an LCA-based methodologyProcedia CIRP, 2024
- Experimental study on the immersion liquid cooling performance of high-power data center serversEnergy, 2024
- A systematic review of Green AIWIREs Data Mining and Knowledge Discovery, 2023
- Energy-aware systems for real-time job scheduling in cloud data centers: A deep reinforcement learning approachComputers and Electrical Engineering, 2022
- Cooling technologies for data centres and telecommunication base stations – A comprehensive reviewJournal of Cleaner Production, 2022
- Assessing ICT global emissions footprint: Trends to 2040 & recommendationsJournal of Cleaner Production, 2018
- Optimizing performance-per-watt on GPUs in high performance computingComputer Science - Research and Development, 2015
- Electrical Grid and Supercomputing Centers: An Investigative Analysis of Emerging Opportunities and ChallengesInformatik Spektrum, 2014