Effective task allocation in fog computing environments using fractional selectivity model

Abstract

In recent scenario, fog computing is a new technology deployed between cloud computing systems and internet of things (IoT) devices to filter out important information from a massive amount of collected IoT data. Cloud computing offers several advantages, but also has the disadvantages of high latency and network congestion, when processing a vast amount of data collected from various devices and sources. For overcoming these problems in fog computing environments, an efficient model is proposed in this article for precise load balancing (LB). The proposed fractional selectivity model significantly handles LB in fog computing by reducing network bandwidth consumption, latency, task-waiting time, and also enhances the quality of experience. The proposed model allocates the required resources by eliminating sleepy, unreferenced, and long-time inactive services. The fractional selectivity model’s performance is investigated on three application scenarios, namely virtual reality (VR) game, electroencephalogram (EEG) healthcare, and toy game. The efficiency of the introduced model is analyzed on the basis of makespan, average resource utilization (ARU), load balancing level (LBL), total cost, delay, and energy consumption. Specifically, in comparison to the traditional task allocation models, the proposed model reduces almost 5 to 15% of the total cost and makespan time.