Mathematical Modeling and Comparative Analysis of PSO and GWO for Efficient Cloud Job Scheduling
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Cloud computing has become a basic part of modern IT services. Job scheduling remains a key technical issue because it influences performance, cost, and resource use. Traditional static and heuristic schedulers often struggle with the dynamic, heterogeneous, and multi-objective nature of current cloud environments. This has encouraged the use of metaheuristic methods. This study presents a systematic empirical comparison of two algorithms (Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO)) for cloud job scheduling. Both algorithms were implemented in a Cloud Sim simulation environment. We used Planet Lab-style workload traces and EC2-like heterogeneous VM catalogs to support repeatable experiments. The evaluation considered Make span, Flowtime, Total Cost, and Average CPU Utilization in a baseline setup (512 tasks, 8 VMs). In the other tests, the number of tasks was increased to 2,000. Under these conditions, the PSO results were more acceptable. Compared with GWO, mean completion time was reduced by about 39.11%. Average CPU utilization increased by 16.21%. Total cost showed a modest decrease of 4.11%. PSO generally was able to produce schedules within fewer iterations. On the other hand, GWO showed a different pattern. When larger populations or longer runtimes were available, it appeared more suitable for deeper exploration. Overall, the study provides a reproducible methodology, a documented data pipeline, and a set of empirical benchmarks.
Fields of Study & Subjects
Cloud Computing and Resource Management
Metaheuristic Optimization Algorithms Research
Distributed and Parallel Computing Systems
Publication details
- Journal / source
- Journal of Al-Qadisiyah for Computer Science and Mathematics
- Volume
- 18
- Issue
- 1
- Publication date
- 30 March 2026
- Language
- EN