ENERGY MANAGEMENT OF CLOUD DATA CENTER USING NEURAL NETWORKS

Worldwide, the number of data centres has increased dramatically as a result of the fast spread of cloud computing, and each facility is now mostly responsible for energy usage. With the goal of improving resource utilisation and lowering power consumption and related costs, the optimisation of energy usage in cloud data centres has emerged as a critical field of academic research. Traditional methods of managing energy have drawbacks and could not offer all-encompassing answers to problems with energy regulation. Recent studies have shown that systems based on machine learning can provide a more effective strategy for controlling energy usage in cloud data centres. This article provides a thorough analysis of the prior research on the use of machine learning approaches to regulate the energy consumption of cloud data centres. A variety of machine learning approaches put out in the body of existing literature are examined in the current study. The aforementioned strategies also include the deployment of virtual machines, server consolidation, and dynamic voltage and frequency scaling. The limitations of using machine learning-based methodology for energy management in cloud data centres are examined in this paper, along with the difficulties that result from the use of such approaches. According to the study's findings, machine learning techniques may be used to effectively address the problems associated with energy management. To maximise resource utilisation, reduce energy consumption, and alleviate related costs, more research and development are required. A more efficient and environmentally friendly cloud computing ecosystem will eventually result from the adoption of these approaches