An intelligent offloading and resource allocation using Fuzzy-based HHGA algorithm for IoT applications

The need for intelligence in today’s era has tremendously increased the demand for Internet of Things (IoT) devices implanted to collect and process diverse data. Cloud computing offers a plethora of services to computationally constrained internet of things devices, but latency degrades the performance of real-time applications. A few other computing paradigms have been developed through the years to overcome this limitation of IoT devices. Fog computing, one of these paradigms, comes into picture as a backbone and offers services to applications which are required to be processed within a deadline. However, addressing challenges such as heterogeneity, offloading mechanisms, resource allocation, and complexity is crucial. This paper presents a framework for intelligent offloading mechanisms and an efficient resource allocation that results in improving the quality of service (QoS) parameters in an integrated cloud-fog-IoT environment. The proposed Fuzzy based Harris Hawks -Genetic Algorithm (HHGA) applies fuzzy-based logic to offload tasks to respective paradigms (cloud or fog), where the upcoming IoT request will be executed. In addition, the Fuzzy-based HHGA algorithm is developed by combining conventional Harris Hawks Optimization (HHO) and Genetic Algorithm (GA) to improve the exploration and exploitation. The proposed algorithm is eventually integrated with the present framework to search for the optimal resources for upcoming requests and reduce the service cost, time, and energy consumption. The experiments are conducted and consecutively the performance of the proposed framework is evaluated. The results demonstrate that the proposed algorithm outperforms Harris Hawks Optimization by 16.95%, Genetic Algorithm by 38.23% and Particle Swarm Optimization (PSO) by 23.09%.