Advanced Grey Wolf Optimizer for Multi-Objective Workflow Scheduling in Fog-Cloud Environments

Abstract

In a world where digitization is ubiquitous in many business sectors, optimizing computing resources is essential for ensuring efficient and sustainable production. Faced with the explosion of data generated by connected devices, requiring real-time processing, the fog-cloud environment offers a promising solution for improving resource management and decision-making. Workflow scheduling in fog-cloud environments is both a popular and complex topic, tackled by several researchers with different objectives. To overcome this NP-complete problem, several methods have been developed. This paper presents an Adaptive Multi-Objective Grey Wolf Optimizer algorithm (AMOGWO) for efficient workflow scheduling in hybrid fog-cloud environments. AMOGWO simultaneously optimizes three critical metrics—makespan, cost and latency-using Pareto dominance principles. Experimental evaluations conducted with Fog-WorkflowSim show the effectiveness of AMOGWO compared with the Multi-objective Monarch Butterfly Optimization (MO-MBO) algorithm and the Multi-Objective Grey Wolf Optimizer (MOGWO) algorithm in terms of solution quality and achieving a good trade-off.