A Modified Jellyfish Search Algorithm for Task Scheduling in Fog-Cloud Systems

Abstract

Integration of fog and cloud has become increasingly important in the age of IoT, where everything is connected to the Internet. The cloud-only models face many challenges when serving the requests from IoT devices due to several factors such as latency, network congestion, data privacy, and security. Despite the popularity and numerous advantages of hybrid models, task scheduling is still an unsolvable multiobjective optimization problem. This research uses an improved bioinspired jellyfish search algorithm to solve the nonlinear np-hard task scheduling optimization problem. The work proposes a multiobjective improved jellyfish search (MOIJS) framework using a multiobjective adaptation function to minimize the make-span, cost, and power consumption that benefit customers and providers by considering the expenses associated with execution and power consumption. The performance of MOIJS is evaluated by comparing it with the discrete nondominated sorting genetic algorithm II using a MATLAB simulator. The experimental outcomes demonstrate the proposed work's efficacy in reducing the make-span, cost, and energy in cloud-fog environments in different batches of tasks.