A Comprehensive Study of Energy Efficiency in Cloud Computing Using Optimization Algorithms

Cloud Computing (CC) is a fundamental technology for distributed networks, enabling the execution of programs across multiple networked computers simultaneously. The major challenge in cloud computing (CC) is optimizing energy utilization, which involves improving the ratio of power consumed by the entire facility to the power consumed, specifically by utilizing IT resources. Energy efficiency in cloud computing includes various techniques like Load balancing, task scheduling, and Resource Allocation (RA). In CC, load balancing is a process which distributes the workloads and traffic to make sure that no single server or system is overloaded or under-loaded. Task scheduling in cloud computing is utilized to allocate tasks to the best suitable resource for execution. The RA in CC is process of allocating available resources to required cloud applications over internet and starving the services when allocation is not controlled correctly. In this study, various methodologies like Modified Multi-Objective Harris Hawk Optimization (MMHHO), Deep Reinforcement Learning with Parallel Particle Swarm Optimization (DRLPPSO), Convolutional Neural Network (CNN) - Modified Particle Swarm Optimization (MPSO), and so on, are used to ensure energy efficiency in cloud computing. These algorithms attain high resource utilization and decreases the response time and cost of maintenance.