An Intelligent Completion Time Aware Task Scheduling (CTATS) Algorithm for Fog Computing Environment for Delay Sensitive IoT Applications

Fog computing also known as Edge computing is emerging as an attention gaining computing technology for providing Cloud services to IoT applications which requires instantaneous and real-time responses. In absence of an efficient and systematic task scheduling algorithm, Fog with limited computing power and resources may fail to handle high traffic and hence may result in delays. Although many studies have been carried out and many task scheduling models has been proposed by researchers pertaining to different application areas, many of these models have some inherent drawbacks like increase in response time and execution time, as well as migration delays which hinders these models from being the most competent ones particularly the case of delay sensitive IoT applications. Hence, being mindful of the mentioned issues and challenges, in this paper, an efficient Completion Time Aware Task Scheduling (CTATS) algorithm has been proposed at Fog Computing layer. We have implemented reinforcement learning (RL), a machine learning strategy that is environment-adaptive, to enhance performance with respect to predict the suitable VM for a delay sensitive task and for this we have used our own dataset. To check the efficacy of the proposed model, the simulation has been carried out using CloudSim simulator and the obtained results is compared with the existing models for the comparison of average response time and overall completion time. Obtained results shows the superior efficacy of CTATS over the QBT and Prioritized Scheduled Mechanism.