HSSIW: Hybrid Squirrel Search and Invasive Weed Based Cost-Makespan Task Scheduling for Fog-Cloud Environment

Abstract

The large-scale development of Internet of Things devices emerged a new computing environment called fog computing to reduce the makespan and cost spent on the cloud devices as a result of distant communication. However, unless the appropriate assignment of tasks is strictly allocated on an available resource of fog nodes, it results in wastage of resources and unachievable quality of service. In this paper, the balance of the most common conflicting objectives in task scheduling that is makespan and cost for the distributed fog-cloud environment is investigated. A novel hybrid squirrel search and invasive weed (HSSIW) algorithm is adapted to assign generated tasks from the Internet of Things(IoT) devices at appropriate fog and cloud nodes so that reduction in cost and makespan is assured. The proposed algorithm has been compared with three related state-of-the algorithms such as genetic algorithm (GA), particle swarm optimization algorithm (PSO), and squirrel search algorithm(SS). The experiment conducted on Cloudsim shows that the proposed algorithm reduces makespan 18% better than classic algorithms such as First Come First Serve(FCFS) and Short Job First(SJF) algorithms. Similarly, it has made a reduction in latency 4 % better than GA and PSO with optimal cost.