NFV-compliant Traffic Monitoring and Anomaly Detection based on Dispersed Vantage Points in Shared Network Infrastructures

Abstract

In this paper we propose a monitoring architecture based on dispersed vantage points in networking infrastructures. Our framework offers on-demand network monitoring data and related analytics to users and/or organizations, considered as tenants of shared infrastructures. Measurements are collected from monitoring agents scattered in a legacy production network, offering authorized users (tenants and administrators) better visibility of events. Adhering to the NFV principles we implemented, deployed and orchestrated management-plane monitoring and anomaly detection services using Docker containers. Thus, our architecture enables users to process, analyze and visualize customized network monitoring data. The proposed architecture was shown to be well-suited for anomaly detection schemas by considering measurements gathered from centralized and localized vantage points within a network.