Delay Aware Global Information Processing for Offloading Computation at Core Level Using Nature-Inspired Algorithm

Abstract

Introduction: Nowadays, fog computing has emerged as a promising solution for handling the prompt processing of tasks in Internet of Things (IoT)-based applications. One of the key advantages of fog computing is that it reduces service completion time by offloading tasks from IoT devices to the fog server. Therefore, scheduling of tasks becomes vital, where emergency and non-emergency tasks can be prioritized to offload data to the nearby fog servers, which improves the Quality of Service (QoS). Due to the dynamic nature of the IoT environment, traffic load varies over time, making it difficult to select the optimal fog server for task offloading. 

Objectives: This research introduces a novel task offloading for the Fog Cloud scenario using the improved Coati Optimization Algorithm based on Genetic Operators (ICOA-GO) algorithm.

Methods: Initially, the Delay aware Four Queue model with Fuzzy logic (DFQM-Fuzz) is designed to queue incoming tasks into four different priorities. In the DFQM-Fuzz,  Highly urgent (HU) and Urgent (U) tasks are considered as first and second priorities and those tasks are offloaded through the Fog Server. The Non Urgent (NU) and No Deadline (ND) tasks are offloaded through the Cloud Server. Furthermore, when there is no resource to offload through the Fog Server, the task is offloaded through the Cloud Server.

Results: The proposed improved Coati Optimization Algorithm based on Genetic Operators (ICOA-GO) algorithm optimizes the selection of Cloud and Fog Servers. The analysis based on Energy Utilization, Task Rejection Ratio, and Makespan yielded values of 93.589, 1, and 1.96073.

Conclusions: By integrating the DFQM-Fuzz and ICOA-GO improves task prioritizing and offloading efficiency while reducing execution time, energy consumption, and cost.