Virtual Machine Classification-based Approach to Enhanced Workload Balancing for Cloud Computing Applications

Despite the many researches that have been conducted in the field of Cloud computing, it is still facing some issues and challenges, such as load balancing which still needs more optimizing methodologies and models to improve performance and achieve high user satisfaction. In this paper, Machine Learning Technique, which is classification, is used to make groups of VMs based on their CPU and RAM utilization, as well as to classify user jobs/tasks into different groups based on their sizes and based on information from log files. The approach arranges virtual machines in groups, and several tasks share the same VM resources. The goal of our proposal is to allow more dynamic resources and to improve the QoS requirements by maximizing the usage of the resources and user satisfaction, such as increasing resource utilization and reducing the number of job rejections.