Machine Learning Centered Energy Optimization In Cloud Computing: A Review

Abstract

The rapid growth of cloud computing has led to a significant increase in energy consumption, which is a major concern for the environment and economy. To address this issue, researchers have proposed various techniques to improve the energy efficiency of cloud computing, including the use of machine learning (ML) algorithms. This research provides a comprehensive review of energy efficiency in cloud computing using ML techniques and extensively compares different ML approaches in terms of the learning model adopted, ML tools used, model strengths and limitations, datasets used, evaluation metrics and performance. The review categorizes existing approaches into Virtual Machine (VM) selection, VM placement, VM migration, and consolidation methods. This review highlights that among the array of ML models, Deep Reinforcement Learning, TensorFlow as a platform, and CloudSim for dataset generation are the most widely adopted in the literature and emerge as the best choices for constructing ML-driven models that optimize energy consumption in cloud computing.