SDN-assisted Service Placement for the IoT-based Systems in Multiple Edge Servers Environment

Abstract

Edge computing is proved to be an effective solution for the Internet of Things (IoT)-based systems. Bringing the resources closer to the end devices has improved the performance of the networks and reduced the load on the cloud. On the other hand, edge computing has some constraints related to the amount of the resources available on the edge servers, which is considered to be limited as compared with the cloud. In this paper, we propose Software-Defined Networking (SDN)-based resources allocation and service placement system in the multi-edge networks that serve multiple IoT applications. In this system, the resources of the edge servers are monitored using the proposed Edge Server Application (ESA) to determine the state of the edge server and, therefore, the acceptable services by each server. Benefiting from the information gathered by ESA, the service offloading decision would be taken by the proposed SDN Non-core Application (SNA) in a way that ensures an efficient load distribution and better resources utilization for the edge servers. A Weighted Aggregated Sum Product Assessment Method (WASPAS) was used to determine the best edge server. The proposed system was compared with a non-SDN system and showed improvement in the performance and the utilization of resources of the edge servers. Furthermore, the request handling time was considerably reduced and settled in constant rates for a different number of devices.