Efficient Big Data Processing in IoT Environments: A Fog–Cloud Offloading Framework with PFOA

Managing the massive volumes of data generated through Internet of Things (IoT) devices is becoming increasingly challenging with their rapid proliferation, particularly regarding power consumption and processing latency. This paper presents an intelligent Fog–Cloud Offloading Framework based on the Piranha Foraging Optimization Algorithm (IOT-FCOF-PFOA) designed to dynamically balance computational workloads between fog and cloud layers. In contrast to traditional offloading strategies, the proposed scheme incorporates an adaptive optimization model that simultaneously reduces energy expenditure and delay and adapts to real-time fluctuations in network load and task complexity. The Piranha Foraging Optimization Algorithm (PFOA) is utilized as a behavior-inspired controller to improve convergence precision and stability in task assignment, allowing for the efficient processing of heterogeneous IoT traffic. The proposed IOT-FCOF-PFOA achieves 19.17%, 19.62%, and 20.25% higher bandwidth and 262s, 210s, and 285s lesser computational time compared with the existing methods, such as Optimal Fog-Cloud Offloading Framework for Big Data Optimization in Heterogeneous IoT Networks (OCOF-BO-IOT), Fog Computing for Energy-Efficient Data Offloading of IoT Applications in Industrial Sensor Networks (FC-DO-IOT-ISN), and the Safe IoT Applications Allocation Framework for Integrated Fog-Cloud Environs (IOT-AF-IFCE).