Hybrid invasive weed optimization with greedy algorithm for an energy and deadline aware scheduling in cloud computing

Abstract

Cloud computing is being envisioned to be the computing paradigm of the next generation primarily for its advantages of on-demand services, risk transference, resource pooling that is independent of location and ubiquitous network access. Service quality is allocated using various resources in the scheduling process. The deadline refers to the time period from task submission until task completion. An algorithm that has good scheduling attempts at keeping the task executed inside the constraint of the deadline. The Genetic Algorithm (GA) is a common metaheuristic that is used often in literature for procuring solutions that are either optimal or near-optimal. The Invasive Weed Optimization (IWO) is an evolutionary algorithm that is population-based with certain interesting specifications like creations of offspring that are based on the levels of fitness of the parents which increases the size of the population and generates new population by making use of the best among parents and the best among off-springs. The Greedy Algorithms will construct an object that is globally best by means of continuously choosing the option that is locally the best. In this work, a hybrid GA with the Greedy Algorithm and a Hybrid IWO with the Greedy Algorithm that has been proposed for the energy and the deadline-aware scheduling in cloud computing.