A Reinforcement Learning Based Framework for Holistic Energy Optimization of Sustainable Cloud Data Centers

The widespread adoption of cloud data centers has led to a rise in energy consumption, with the associated carbon emissions posing a further threat to the environment. Cloud providers are increasingly moving towards sustainable data centers powered by renewable energy sources (RES). The existing approaches fail to efficiently coordinate IT and cooling resources in such data centers due to the intermittent nature of RES and the complexity of state and action spaces among different devices, resulting in poor holistic energy efficiency. In this paper, a reinforcement learning (RL) based framework is proposed to optimize the holistic energy consumption of sustainable cloud data centers. First, a joint prediction method MTL-LSTM is developed to accurately evaluate both energy consumption and thermal status of each physical machine (PM) under different optimization scenarios to improve the state space information of the RL algorithm. Then, this framework designs a novel energy-aware approach named BayesDDQN, which leverages Bayesian optimization to synchronize the adjustments of VM migration and cooling parameter within the hybrid action space of the Double Deep Q-Network (DDQN) for achieving the holistic energy optimization. Moverover, the pre-cooling technology is integrated to further alleviate hotspot by making full use of RES. Experimental results demonstrate that the proposed RL-based framework achieves an average reduction of 2.83% in holistic energy consumption and 4.74% in brown energy, which also reduces cooling energy consumption by 13.48% with minimal occurrences of hotspots. Furthermore, the proposed MTL-LSTM method reduces the root mean square error (RMSE) of energy consumption and inlet temperature predictions by nearly half compared to LSTM and XGBoost.

I. Introduction

The rapid expansion of cloud data centers has brought attention to the issue of energy consumption. In recent times, the energy consumption of a single server rack has surged from less than 3 kW to 30 kW [1]. The main components of energy consumption have also changed. Besides IT devices, around 40% of the energy consumed by these data centers can be attributed to the Computer Room Air Conditioning (CRAC) systems [2]. This trend will also exacerbate the impact on the environment, and it is estimated that cloud data centers will contribute up to 8% of global carbon emissions by 2030 [3]. Therefore, major cloud providers (such as Google and Amazon) are prioritizing the use of renewable energy sources (RES) like solar and wind energy to build sustainable cloud data centers. Although promising, the energy requirements of these data centers cannot always be satisfied due to the intermittent nature of RES, leading to increased reliance on non-renewable (brown) energy sources. In this context, the energy consumption should be further optimized according to the RES availability.