Resource Allocation Technique for Edge Computing Using Grey Wolf Optimization Algorithm

Abstract

The explosion of IoT technology poses new challenges for researchers in the concept of cloud computing, mainly in improving the distribution of services, which need to be provided with greater efficiency and less latency. Therefore, this work seeks to optimize the methodology of resource allocation in Edge Computing, seeking to improve the quality of service (QoS) to the user. For this, it was developed an algorithm for efficient resource allocation using grey wolves optimization technique, named as Resource Allocation Technique for Edge Computing (RATEC). The algorithm adopted the meta-heuristic technique to choose the best Edge when allocating the resources of user equipment (UE). In this work, it was considered that the UEs are composed of processing, storage, time and memory resources. The algorithm uses these resources to calculate the fitness of each Edge and decide which one to allocate, if available. The RATEC has been compared with two other policies and has managed to serve a number most significant of UEs, reducing the number of services refused and presenting a low number of blockages while searching for an Edge.