A comparison of cloud work scheduling algorithms with a Biogeography-based optimisation algorithm based on a Genetic Algorithm

Abstract

A hybrid cloud is a platform which enables the owner of any business house to have certain domestic resources and various different resources to be procured from some external service provider. This model is quite helpful to avoid as well as handle the condition of a "cloud burst", where a private cloud setup is deemed to deal with the extra burden and then immediately has to switch to a public cloud to deal with the situation. Cloud Scheduling tries to achieve the “most eminent” plan or schedule for datacentres to serve the demanded service with minimum communication losses with the least production cost. This research paper incorporated mutation and crossover along with the generic functions of Biogeography-Based Optimization algorithm. This amalgamation gives an efficient Algorithm, namely, Genetic Algorithm-Biogeography-Based Optimization Algorithm for the hybrid Cloud. The scheduling cost and throughput have been evaluated on an open cloud raw dataset using MATLAB. The proposed work is compared with two existing algorithms, namely, Ant Colony Optimization and Biogeography-Based Optimization algorithm.