Hierarchical Multi-Agent Reinforcement Learning for Carbon-Aware AI Data Centers in Power Distribution Systems
2026-07-03 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors study how to reduce the carbon footprint of AI data centers by managing their energy use more efficiently. They propose a system where one main agent assigns AI training jobs to different data centers, while smaller agents at each center schedule tasks to save energy and reduce emissions. Their approach uses data about energy use and carbon intensity to decide when and where to run AI jobs and how to cool the centers more efficiently. They tested their method using a standard power system model to show its potential benefits.
artificial intelligence data centerscarbon emissionsmulti-agent reinforcement learningworkload managementpower distribution systemcarbon intensityschedulingenergy managementcooling systemstransformer model
Authors
Hyunsoo Lee, Panggah Prabawa, Dae-Hyun Choi, Joongheon Kim
Abstract
Eco-friendly energy management for artificial intelligence data centers (AIDCs) is crucial because of the significant increase in energy consumption-induced carbon emissions from AIDCs resulting from the rapid expansion of AI applications. This paper proposes a hierarchical carbon-aware multi-agent reinforcement learning (CA-MARL) framework for robust and efficient operations of AIDCs under uncertainties while ensuring low-carbon operation of power distribution systems. The framework comprises a workload manager (WM) agent and multiple local AIDC agents trained using a multi-agent transformer method, corresponding to a global AIDC aggregator and a local AIDC operator, respectively. Leveraging AIDC operation data along with nodal carbon intensity (NCI) calculated from the carbon emission flow-integrated distribution system operator problem, the WM agent spatially allocates AI training and inference jobs among all AIDCs. Based on the jobs allocated from the WM agent and NCI information, each AIDC agent schedules economical and eco-friendly operations of the AIDC by performing the following tasks: i) temporal shifting of training jobs, ii) spatial allocation of training graphics processing unit (GPU) blocks and inference GPUs within the AIDC, and iii) control of the supply air temperature of the cooling system. The effectiveness of the proposed framework was assessed using an IEEE 33-node power distribution system.