Effect of Cooling Systems on the Energy Efficiency of Data Centers: Machine Learning Optimisation

The number of data centers is increasing rapidly since more people are now using cloud services for data storage and management. Due to this, the total power consumption of the data centers is also increasing. The energy efficiency of the data centers is not very high due to a variety of reasons like heat loss by equipment and power factor issues. This paper attempts to review the existing work around 2015 to 2019 and understand the issues faced by the data centers. The energy usage by the data centers and the effect of the temperatures are reviewed along with the methods of optimisation through Machine Learning (ML) algorithms. Some of the factors affecting the energy consumption of the data centers are the airflow, heat loss, ambient temperature, among others. The gap in the existing research is obtained by identifying the various factors that affect the cooling of the data centers. The effect of the cooling parameters is optimised at different locations of the data centers as per the requirement. Reinforced learning techniques have been seen to be efficient in terms of optimisation. A combination of Support Vector Machine (SVM) and Ant Colony Optimisation (ACO) is suggested as a future scope of this study.