Predicting Public Sector Energy Efficiency with Cloud Data Centre using a Consolidation of Machine Learning

The next generation of data centres along with the ever-increasing demand for power efficiencies in the infrastructure of cloud computing is creating buzz over innovative solutions. Among various techniques towards the optimization of energy costs along with environmentally friendly options, VM (Virtual Machine) consolidation is seen as one of the most viable solutions. Intelligent learning approaches such as machine learning and deep learning have emerged as powerful management tools towards optimization of the resources of a cloud data centre. Since most of these VM migration approaches between the host failed to consider the migration costs with their consideration of total energy consumptions; some include the migration distance and time. We have reviewed more than 40 research paper and discusses different techniques for improving energy efficiency-from SVM (Support Vactor Machine) to Decision trees, Deep learning, and RL (Reinforcement Learning) while comparing and contrasting the approaches used. We focus specifically on RL methodologies and analyse how well they can reduce energy consumption, towards understanding which one has a promising potential in minimizing energy consumption in cloud data centres.