sFlow monitoring system in a disaster-resilient global SDN testbed based on KOREN/APII/TEIN network

Abstract

This paper provides insights into the traffic flow monitoring system of a disaster-resilient SDN testbed which is based on Korea Advanced Research Network (KOREN)/Asia Pacific Information Infrastructure (APII)/Trans-Eurasia Information Network (TEIN) network. This testbed consists of research and education networks of several countries including South Korea, Japan, and the USA. Since, it is a large multi-tenant testbed, it is not possible for the traditional traffic monitoring solutions to provide network-wide visibility. Therefore, we have implemented a new sFlow-based monitoring system that is not only tailored to the requirements of this testbed, but can also provide real-time network-wide visibility. It is carried out by deploying open-source host sFlow agents with a graphite collector which offers a complete, highly scalable monitoring solution. It periodically fetches and other statistics from sFlow agents, and stores them in time-series format. These statistics are then exported to the disaster management server where they are used for further analysis.