Analysis of intelligent offloading and hybrid optimization for low-latency and energy-efficient Fog–Cloud computing in IoT systems

Abstract

A high rate of increment in the internet-based devices generates huge amounts of data every day in almost all aspects of life. These internet-enabled devices don’t have any storage, processing, or subject capacity to handle and store this large quantity of correct and voluminous data. Nevertheless, the current fog computing solutions have been limited to being able to completely reduce latency through poor task scheduling and high communication impediments. The present paper discusses and enhances the low-latency in Fog computing architectures on the Internet of Things (IoT) systems. This work presents a comparative analytical framework called EDGE-LAT for systematically evaluating the existing latency-aware task offloading and resource allocation approaches in IoT–Fog–Cloud environments. The EDGE-LAT analytical framework evaluates representative latency-aware approaches that are based on task offloading, hybrid optimization, communication delay, computational cost, energy consumption, and execution time. The comparative evaluation of three models, MCEETO, PSOSA-LB, and Chaotic HBA-OBL, is done under various workloads. This comparison shows that the EDGE-LAT analytical framework identifies better latency-aware strategies under the evaluated simulation settings. In particular, the network can attain a latency of 11.42 ms, a power consumption of 0.852 kJ, and an execution time of 2.94 s, which proves that it outperforms the others in the management of real-time Internet of Things. Overall, this analytical study highlights the potential of a latency-aware and energy-efficient task offloading strategy for enhancing the future IoT–Fog–Cloud systems.