Leveraging Machine Learning for Enhanced Cloud Computing Load Balancing: A Comprehensive Review

Abstract

Popularized especially throughout the last decade, cloud computing has significantly changed the approach to the use of computational resources. But load balancing still persists as a very important issue of concern to achieve the right balance in order to optimize the various resources. This review paper aims at discussing the various Machine Learning (ML) techniques that can be used for the improvement of load balancing in cloud computing systems. Therefore, the analysis of the current methods is intended to reveal the advantages of the modern approaches to load balancing and put focus on the shortcomings of the traditional algorithms and the contribution of new machine learning approaches. ML techniques offer dynamic load balancing solutions which can handle the variance of the system load. These techniques mainly involve the use of historical data and applying statistical analysis to arrive at the best decisions thus reducing the time response, server usage, and resource utilization. Further, this paper presents how the state-of-the-art machine learning models can be implemented in the context of cloud environments focusing on the complexity and overhead, data protection, and scalability questions. It continues the review of the most advanced machine learning algorithms in the context of load balancing. Adding several examples of the current case studies and experiment results from the published researches of the recent years, the effectiveness of these techniques has been illustrated in the real cloud context. This paper is concluded with the generalization for the future works where the emphasis is placed for the improvement of load balance of machine learning for more demanding and secured cloud computing service.