Elastic urban video surveillance system using edge computing

Abstract

During the past decade, the concepts and applications of Internet of Things (IoT) are pervasively propagated to the academia and industries. The widely distributed IoT devices contribute to building an effective smart urban surveillance system, which manages the regular operations and handles emergencies. The real time monitoring uploads massive amounts of data to the backbone network and requires prompt feedbacks. The recent rapid development of "Edge Computing" (also called "Fog Computing" or Mobile Edge Computing in different literature) aims at pushing the computation and storage resources from the remote data center to the edge of network for reducing the burden of backbone and the computing latency In this paper, we design a three-tier edge computing system architecture to elastically adjust computing capacity and dynamically route data to proper edge servers for the real-time surveillance applications. A system prototype integrating Network Functions Virtualization (NFV) and Software-Defined Networking (SDN) is implemented in an OpenStack based virtualization environment. Moreover, we introduce schemes of resource reallocation and workload balance in urgent situations. Experimental results of the prototype show the great potentials of using edge computing for future large-scale and distributed smart urban surveillance applications.