MOHBA: Multi-objective Honey Badger Algorithm for workflow scheduling in heterogeneous Cloud–Fog-IoT networks

Nowadays, the Internet of Things (IoT) is used for the purpose of gathering data through sensors as well as for storing and processing data. Due to the inherent limitations in the processing and computing capabilities of IoT devices, the inclination towards the integration of cloud computing (CC) with IoT systems is growing rapidly. The CC is capable of efficiently handling substantial volumes of data at remarkable speeds. However, the data owners are required to upload their data to the cloud, and the transmission of such extensive data requires a significant amount of bandwidth. Moreover, the existence of latency and jitter arises as a consequence of the significant geographical separation between IoT devices and the cloud. Hence, fog computing (FC) is utilized in proximity to IoT devices so that the extent of transmission delay is diminished. Both CC and FC are employed in collaboration to improve the performance of the IoT system. A wide range of workflow scheduling optimization techniques were proposed for both CC and FC environments. However, in heterogeneous remote computing systems, major challenges encountered are related to execution time, energy efficiency, latency, cost, and load balancing. In this study, the Multi-objective Honey Badger Algorithm (MOHBA) is proposed for optimizing workflow schedules aimed at optimizing makespan, energy consumption, and overall cost, which hold significant importance for real-time systems. FogWorkflowSim is used for simulating where HBA outperforms other existing algorithms like Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and Artificial Algae Algorithm (AAA).