Google Research

Data Center Cooling using Model-predictive Control

Proceedings of the Thirty-second Conference on Neural Information Processing Systems (NeurIPS-18), Montreal, QC (2018), pp. 3818-3827

Abstract

Despite the impressive advances in reinforcement learning (RL) algorithms, their deployment to real-world physical systems is often complicated by unexpected events and the potential for expensive failures. In this paper we describe an application of RL “in the wild” to the task of regulating temperatures and airflow inside a large-scale data center (DC). Adopting a data-driven model-based approach, we demonstrate that an RL agent is able to effectively and safely regulate conditions inside a server floor in just a few hours, while improving operational efficiency relative to existing controllers.

Research Areas

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work