Improving flexibility in cloud computing using optimal multipurpose particle swarm algorithm with auction rules

Abstract

ABSTRACT Using exact algorithms in complex environments with multi-objective criteria cannot be considered an appropriate method. For this reason, nature-inspired meta-heuristic methods have been proposed to solve problems such as flexibility in the cloud computing environment. In this research, we use the particle swarm optimisation algorithm and auction approach for two-way load balancing. A cloud computing centre comprises several physical machines called data centres, and data centres consume a lot of energy. To increase efficiency, there is a need to increase flexibility in these systems. The researchers proposed Various methods such as static and dynamic management of virtual machines and automatic processor frequency change systems to reduce energy consumption. In this article, we presented an algorithm based on under-load hosts and particle swarm optimisation and mass movement of virtual machines according to the auction rule to reduce energy as the dynamic management of cloud resources. The proposed method has been evaluated by the cloudsim simulator with PlanetLab real dataset. The experimental results show that, in comparison to the other algorithms, energy consumption has improved and SLA violation has decreased. As well as migration of VMs decreased significantly.