In recent years, cloud computing has been widely adopted over the Internet due to the numerous advantages and services it offers to users. The resources in cloud computing are based on virtualization, which makes them dynamic and prone to frequent changes. These resources are utilized by cloud services or user applications in order to respond to user requests. The smallest unit of a user application, known as a job, requires exclusive access to certain resources which would enable its execution. Assigning jobs to cloud nodes (a set of computational resources) is a NP-complete problem, commonly referred to as job scheduling in the cloud. The aim of this paper is to propose an integer programming model for cloud job scheduling and to find an optimal or near-optimal solution by using two algorithms, namely a genetic algorithm called GAJSC, and a particle swarm optimization (PSO) algorithm. This study concludes with a comparison of the performance of these two approaches against certain traditional baseline algorithms, including the First-Come-First-Serve (FCFS) and Shortest Job First (SJF) algorithms in terms of the obtained makespan and their scalability.
Cloud computing, Job scheduling, Optimization, Genetic algorithm, PSO, Cloudsim Plus, Virtualization, Makespan.
Abdelbasset BARKAT, Zohair TAHRI, Derya YILTAS-KAPLAN, "A Comparative Study of Job Scheduling in Cloud Computing", Studies in Informatics and Control, ISSN 1220-1766, vol. 34(3), pp. 81-90, 2025. https://doi.org/10.24846/v34i3y202508
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).