Workflow task scheduling in a cloud-fog environment: a hybrid PSO-GOA approach

作     者：Bansal, Sumit Singla, Bhim Sain Aggarwal, Himanshu 

作者机构：Thapar Inst Engn & Technol Dept Comp Sci & Engn Patiala 147004 Punjab India Punjabi Univ Dept Comp Sci Neighbourhood Campus Ralla 151510 Punjab India Punjabi Univ Dept Comp Sci & Engn Patiala 147002 Punjab India 

出 版 物：《INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT》 (Intl. J. Syst. Assur. Eng. Manage.)

年 卷 期：2025年

页      面：1-23页

核心收录：

主　　题：Cloud Fog Scheduling Workflow Hybridization Optimization Grasshopper Optimization Algorithm 

摘      要：Cloud-fog computing frameworks represent emerging paradigms designed to enhance existing Internet of Things (IoT) architectures. In these frameworks, task scheduling is crucial for optimizing task allocation and execution within the cloud-fog computing environment. Finding an optimal algorithm for workflow scheduling poses a significant challenge due to the complex nature and variable aspects of the tasks and resources involved. Metaheuristic algorithms are the ones which can overcome this problem and also offers a variety according to the nature of problem. However, they frequently encounter issues such as getting trapped in local optima, which delays their ability to attain the global optimal solution. So, as to eliminate this issue, we employed a hybrid approach known as the Hybrid Particle Swarm Optimization Algorithm-Grasshopper Optimization Algorithm (HPSO-GOA). By leveraging the characteristics of both PSO and the GOA, the proposed algorithm adeptly addresses the issue of becoming trapped in local optima. The objective of this work is to minimize the Total Execution Time (TET), Total Execution Cost (TEC), and Energy Consumption (EC). Based on the evaluation metrics, our Multiobjective optimization algorithm outperforms in terms of TET achieving an overall average reduction of 6.69% for PSO and 20.60% for the GOA and 1.65% for Hybrid Genetic Algorithm-Particle Swarm Optimization (HGA-PSO). When compared for TEC and EC it outperforms PSO by 3.29%, 10.84%, GOA by 16.46%, 17.85% and HGA-PSO by 4.40%, 3.27% respectively. To evaluate the effectiveness of the proposed algorithm, we conducted comparative analyses with state-of-the-art algorithms (PSO, GOA, HGA-PSO) across five diverse scientific workflows. These comparative analyses and statistical analysis (Wilcoxon and Friedman) highlighted the effectiveness of the HPSO-GO algorithm in enhancing workflow scheduling performance.