Machine Learning Solutions for Load Balancing in Cloud Resource Management: A Review

Cloud computing has rapidly gained popularity as a means of delivering utility computing services over the Internet It provides access to a shared repository of digital resources housed in energy-intensive data centers, making it a viable option for accommodating diverse workloads of information and communication services. Efficient resource management has become a crucial goal for cloud providers, leading them to explore data-driven strategies utilizing machine learning techniques. Machine learning algorithms are now widely applied in resource management tasks, including load balancing, Virtual Machine consolidation, efficient resource allocation, and energy optimization. This review paper specifically focuses on machine learning-based load-balancing solutions in cloud resource management. It begins by introducing the fundamental principles of Cloud computing, emphasizing the need for resource management and the importance of machine learning in this context. The paper further explores the role of load balancing in resource management and discusses the associated challenges. The review proceeds chronologically from 2015 to 2022, presenting a comprehensive overview of advancements and innovations in load-balancing techniques. A comparative evaluation, in tabular format, assesses the benefits and limitations of each approach. In conclusion, the paper discusses open research challenges identified in the literature and suggests potential future directions for further research in this field.