This post is also available in:
עברית (Hebrew)
As artificial intelligence (AI) continues to advance, the energy consumption of data centers, which support AI computations, has been rising sharply. A recent study by Lawrence Berkeley National Laboratory revealed that U.S. data centers consumed 4.4% of the nation’s electricity in 2023, with consumption expected to triple by 2028. This surge is largely driven by the widespread use of AI, creating significant economic and environmental challenges.
In response, researchers at the National Renewable Energy Laboratory (NREL) have been investigating ways to make computing more energy-efficient. Their latest contribution, A Beginner’s Guide to Power and Energy Measurement and Estimation, developed in collaboration with Intel, aims to equip AI professionals with the tools and knowledge to measure and manage energy usage effectively.
According to TechXplore, the guide is designed for machine learning developers and software engineers, providing them with a structured approach to energy measurement across different computational levels. It addresses key questions and offers practical advice on how to select the right tools to analyze energy consumption at the system, application, and code levels. The goal is to foster more sustainable decision-making in computing, allowing developers to balance performance and energy efficiency.
Hilary Egan, data scientist and lead author of the report, emphasized the importance of collaboration in addressing AI’s energy challenge. “Through this guide, we wanted to provide AI professionals with an introduction to energy estimation that opens the door to more sustainable decision-making in computing”, she said, according to TechXplore.
The guide is a key product of NREL’s Green Computing Catalyzer, launched in 2022 as part of the Joint Institute for Strategic Energy Analysis (JISEA). The initiative brings together industry experts, researchers, and universities to explore ways to reduce the energy footprint of machine learning and AI systems.
The guide provides developers with the tools necessary to make intelligent energy decisions—an essential step toward reducing AI’s environmental impact. This collaboration between NREL and Intel represents a crucial step in making AI more energy-efficient and sustainable, addressing one of the key challenges of modern computing. By providing a clear framework for energy measurement, the guide helps developers contribute to a more sustainable future for AI and data processing.
