An autonomic offloading and resource allocation technique for IoT applications in edge computing

An autonomic offloading and resource allocation technique for IoT applications in edge computing

연구 분야: Networking

학회: The Journal of Supercomputing

Due to explosion of Internet of Things (IoT), the attention of numerous researchers is moved towards the service distribution that requires reduced latency and higher efficiency. As a result, the concept of edge paradigm has been introduced which takes computation and storage in the vicinity of edge nodes. To ensure optimal performance, cost-effectiveness and scalability, offloading decision plays a vital role to decide the efficacy of quality of service (QoS) parameters. Hence, a competent technique is proposed to address the offloading in dynamic environment, uncertainty and complex optimization problem. Large number of metaheuristic approaches has been developed for efficient offloading and resource allocation, but unable to find optimum solution due to inappropriate balanced between exploration and exploitation, slow convergence and trap in local minima etc. We have proposed a hybrid metaheuristic algorithm that is the integration of whale optimization algorithm (WOA) and grey wolf optimization (GWO) with goals to optimize the allocation of resources in edge computing to enhance the QoS. The ability of proposed work is to search the optimal resource for offloading and allocation to improve the delay, cost and energy consumption. The experimental results show that this work reduces the total delay by approximately up to 41%, cost by up to 33% and energy consumption up to 39% in comparison with baseline algorithms.