An Experimental Study on Inter-domain Routing Dynamics using IP-level Path Traces
Commercial relationships between Autonomous Systems (ASes) regulate the inter-domain routing dynamics. Identifying these relationships and properly interpreting them can help understand inter-domain routing dynamics and explore routing anomalies where ASes are not following standard transit policies. Most studies in this domain use Border Gateway Protocol (BGP) data to study inter-domain routing dynamics. In this study, we use IP-level topology data to complement these studies. More specifically, we use traceroute tool to collect end-to-end path traces. Then, we map our data to an AS-level topology map. We do this mainly by using the existing BGP based topology maps and AS-level neighborhood data to come up with several new heuristics to identify false AS-level links and missing AS hops in our dataset. In addition, we infer relationships for those AS-level links in our IP-level data that do not appear in the existing BGP based topology maps. We evaluate the accuracy of our heuristics by comparing their findings to that of the existing state-of-the-art techniques. Finally, we use AS-level paths along with AS relationships dataset to infer potential inter-domain routing anomalies and classify them according to their suspected causes.