Particle Swarm Algorithm Parameters Analysis for Scheduling Virtual Machines in Cloud Computing

Abstract

The computational demand of recent years has made a new computational paradigm become extremely necessary to meet the demand for resources. Cloud computing has been widely used and is a reality in all sectors that demand computational use allied with security and with ease of management. Gigantic data centers were created to meet ever-increasing demand. Processing, memory and storage are delivered to end customers who do not have the energy, cooling, hardware, software, licensing and management concerns, paying only for what they really need. Considering that the user requests resources to perform a certain task, it is necessary to create efficient mechanisms of allocation of resources and fair collection metrics. In this work a review is made of concepts of cloud computing and resource scheduling and analyze two scheduling algorithms that use particle swarm. Finally, the particle swarm algorithm is implemented to make analysis of the best parameter configuration to meet the demand for virtual machine-allocation in cloud computing. The amount of CPU, memory and disk is considered for calculation.