Reference Hub2
Solving Task Scheduling Problem in the Cloud Using a Hybrid Particle Swarm Optimization Approach

Solving Task Scheduling Problem in the Cloud Using a Hybrid Particle Swarm Optimization Approach

Salmi Cheikh, Jessie J. Walker
Copyright: © 2022 |Volume: 13 |Issue: 1 |Pages: 25
ISSN: 1947-8283|EISSN: 1947-8291|EISBN13: 9781799885405|DOI: 10.4018/IJAMC.2022010105
Cite Article Cite Article

MLA

Cheikh, Salmi, and Jessie J. Walker. "Solving Task Scheduling Problem in the Cloud Using a Hybrid Particle Swarm Optimization Approach." IJAMC vol.13, no.1 2022: pp.1-25. http://doi.org/10.4018/IJAMC.2022010105

APA

Cheikh, S. & Walker, J. J. (2022). Solving Task Scheduling Problem in the Cloud Using a Hybrid Particle Swarm Optimization Approach. International Journal of Applied Metaheuristic Computing (IJAMC), 13(1), 1-25. http://doi.org/10.4018/IJAMC.2022010105

Chicago

Cheikh, Salmi, and Jessie J. Walker. "Solving Task Scheduling Problem in the Cloud Using a Hybrid Particle Swarm Optimization Approach," International Journal of Applied Metaheuristic Computing (IJAMC) 13, no.1: 1-25. http://doi.org/10.4018/IJAMC.2022010105

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Synergistic confluence of pervasive sensing, computing, and networking is generating heterogeneous data at unprecedented scale and complexity. Cloud computing has emergered in the last two decades as a unique storage and computing resource to support a diverse assortment of applications. Numerous organizations are migrating to the cloud to store and process their information. When the cloud infrastructures and resources are insufficient to satisfy end-users requests, scheduling mechanisms are required. Task scheduling, especially in a distributed and heterogeneous system is an NP-hard problem since various task parameters must be considered for an appropriate scheduling. In this paper we propose a hybrid PSO and extremal optimization-based approach to resolve task scheduling in the cloud. The algorithm optimizes makespan which is an important criterion to schedule a number of tasks on different Virtual Machines. Experiments on synthetic and real-life workloads show the capability of the method to successfully schedule task and outperforms many known methods of the state of the art.