Hierarchical Multi-Agent Framework for Carbon-Efficient Liquid-Cooled Data Center Clusters
Soumyendu Sarkar, Avisek Naug, Antonio Guillen, Vineet Gundecha, Ricardo Luna Gutiérrez, Sahand Ghorbanpour, Sajad Mousavi, Ashwin Ramesh Babu
Abstract
Reducing the environmental impact of cloud computing requires efficient workload distribution across geographically dispersed Data Center Clusters (DCCs) and simultaneously optimizing liquid and air (HVAC) cooling with time shift of workloads within individual data centers (DC). This paper introduces Green-DCC, which proposes a Reinforcement Learning (RL) based hierarchical controller to optimize both workload and liquid cooling dynamically in a DCC. By incorporating factors like weather, carbon intensity, and resource availability, Green-DCC addresses realistic constraints and interdependencies. We demonstrate how the system optimizes multiple data centers synchronously, enabling the scope of digital twins, and compare the performance of various RL approaches based on carbon emissions and sustainability metrics while also offering a framework and benchmark simulation for broader ML research in sustainability.
BibTeX
@article{Sarkar_Naug_Guillen_Gundecha_Luna Gutiérrez_Ghorbanpour_Mousavi_Ramesh Babu_Rengarajan_Bash_2025, title={Hierarchical Multi-Agent Framework for Carbon-Efficient Liquid-Cooled Data Center Clusters}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/35370}, DOI={10.1609/aaai.v39i28.35370}, abstractNote={Reducing the environmental impact of cloud computing requires efficient workload distribution across geographically dispersed Data Center Clusters (DCCs) and simultaneously optimizing liquid and air (HVAC) cooling with time shift of workloads within individual data centers (DC). This paper introduces Green-DCC, which proposes a Reinforcement Learning (RL) based hierarchical controller to optimize both workload and liquid cooling dynamically in a DCC. By incorporating factors like weather, carbon intensity, and resource availability, Green-DCC addresses realistic constraints and interdependencies. We demonstrate how the system optimizes multiple data centers synchronously, enabling the scope of digital twins, and compare the performance of various RL approaches based on carbon emissions and sustainability metrics while also offering a framework and benchmark simulation for broader ML research in sustainability.}, number={28}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Sarkar, Soumyendu and Naug, Avisek and Guillen, Antonio and Gundecha, Vineet and Luna Gutiérrez, Ricardo and Ghorbanpour, Sahand and Mousavi, Sajad and Ramesh Babu, Ashwin and Rengarajan, Desik and Bash, Cullen}, year={2025}, month={Apr.}, pages={29694-29696} }