Energy-Aware Metaheuristic Algorithm for Industrial-Internet-of-Things Task Scheduling Problems in Fog Computing Applications

In Industrial-Internet-of-Things (IIoT) applications, fog computing (FC) has soared as a means to improve the Quality of Services (QoSs) provided to users through cloud computing, which has become overwhelmed by the massive flow of data. Transmitting all these amounts of data to the cloud and coming back with a response can cause high latency and requires high network bandwidth. The availability of sustainable energy sources for FC servers is one of the difficulties that the service providers can face in IIoT applications. The most important factor contributing to energy consumption on fog servers is task scheduling. In this article, we suggest an energy-aware metaheuristic algorithm based on a Harris Hawks optimization algorithm based on a local search strategy (HHOLS) for task scheduling in FC (TSFC) to improve the QoSs provided to the users in IIoT applications. First, we describe the high virtualized layered FC model taking into account its heterogeneous architecture. The normalization and scaling phase aids the standard Harris hawks algorithm to solve the TSFC, which is discrete. Moreover, the swap mutation ameliorates the quality of the solutions due to its ability to balance the workloads among all virtual machines. For further improvements, a local search strategy is integrated with HHOLS. We compare HHOLS with other metaheuristics using various performance metrics, such as energy consumption, makespan, cost, flow time, and emission rate of carbon dioxide. The proposed algorithm gives superior results in comparison with other algorithms.

Automated summary

This article proposes an energy-aware metaheuristic algorithm based on a Harris Hawks optimization algorithm for task scheduling in FC to improve the QoS provided to users in IIoT applications. Through comparisons with other metaheuristics, the proposed algorithm achieves superior results in terms of energy consumption, makespan, cost, flow time, and carbon dioxide emission rate.