Edge Computing Dynamic Resource Management: Tradeoffs Between Security and Application QoE

Abstract

With the advancement of the 5G network and Internet of Things (IoT) devices, Multi-access Edge Computing (MEC) proposed by ETSI provides multiple devices to access with low latency through heterogeneous networks such as smart factories and vehicular networks. In addition, video streaming and online gaming have become more popular and consume more than half of the traffic on the internet. Thus, there will be more edge servers deployed on the edge of the network for offloading the core network. However, the edge server is more vulnerable because of its proximity to the user equipment (UE). Attackers can quickly launch distributed denial-of-service (DDoS) attacks with plenty of infected IoT devices. In this paper, we propose Tradeoffs Between Security and Application QoE (TBSA) system to solve the security and resource management problems on the edge server. First, we deploy video streaming, online gaming, and network security applications on the edge server. We use Intrusion Detection and Protection Services (IDPS) to perform DDoS mitigation and design resource allocation algorithm to allocate the computing resources to the edge computing applications. Then, we compare different attack rates in the user scenarios and analyze multiple models under the resource limit condition. The experiments show that we can improve the Quality-of-Experience (QoE) of applications by edge computing resources management.