Analytical Workload Allocation for Optimized Power Consumption and Delay in Fog-Cloud Networks Using Particle Swarm Optimization Algorithm

Abstract

Fog Computing caters to the immediate requirement of processing and storage to serve the application running on physical devices attached to the Internet of Things. However, edge devices are restricted by limited processing power and storage. Thus, the voluminous data and the velocity with which the devices produce this data must be judiciously distributed between the fog device and data centers in the cloud system. Such an arrangement ensures that the overall quality of service and quality of experience of end-users are maintained. The paper proposes an algorithm to allocate jobs that utilizes branch-bound technique and particle swarm optimization, a tradeoff between power consumption and task execution time over fog-cloud networks. The paper addresses the twin challenge of reducing the delay period or latency for communication and reducing the power consumption of the fog devices and machines deployed in the data centers. The paper aims to optimize workload allocation among various system components, including fog devices, machines in the data center, and the communication network between them. The graphical findings demonstrate both power consumption and delay have been reduced by employing the suggested algorithm in a fog-cloud scenario.
Copyright © 2023 Praise Worthy Prize - All rights reserved.