A third generation genetic algorithm NSGAIII for task scheduling in cloud computing
Imene Latreche, Sihem Slatnia, Okba Kazar, Mohamed Batouche
Abstract
Task scheduling in the cloud is perceived as a difficult Multi-objective optimization problem. It refers to the assignment of user tasks on the available cloud virtual machines. This problem can be solved effectively by combining two or more approaches for improving task execution and increasing the use of resources. In this article, a third-generation Multi-objective optimization method called Non-dominated Sorting Genetic Algorithm (NSGA-III) was used for the first time to our knowledge to schedule a set of user tasks on a set of available virtual machines (VMs) in the cloud based on a new Multi-objective adaptation function to minimize the runtime (TE), the power consumption (CE), and the cost (cout). Furthermore, the performance of NSGAIII was compared with the performance of its previous version, Non-dominated Sorting Genetic Algorithm (NSGAII) where NSGAIII results outperform the results of NSGAII. The experimental results of the proposed method are encouraging, as they are used to show their effectiveness in solving such problems.