Multi-Objective Task Scheduling in the Cloud Environment Using Reinforcement Learning and Whale Optimization Algorithm

Cloud Computing (CC) enables organizations and individuals to alter and revolutionize the procedure by which people obtain and utilize resources. Due to the growth of the CC, the complexity increases and the cloud systems demand an effective solution for optimal resource allocation. However, effective utilization of resources in CC systems remains a significant problem due to their scale, heterogeneity, and dynamic nature. Artificial Intelligence (AI) has grown up as an effective means to improve management of resources effectively. Task scheduling is one of the major aspects in resource allocation and this study focuses on dynamic task scheduling algorithms to effectively allocate the task to the appropriate Virtual Machines (VMs). This study examines and suggests the reinforcement learning approach with a whale optimization algorithm (RL-WOA) for efficient task allocation in the cloud environment and its performance in terms of efficiency measures in terms of QoS metrics, makespan, CPU utilization and memory usage. The implementation performed using a CloudSim simulator for demonstrating the efficiency of the suggested model.