A Survey of Techniques for Internet Topology Discovery

Capturing an accurate view of the Internet topology is of great interest to the networking research community as it has many uses ranging from the design and evaluation of new protocols and services to the vulnerability analysis of the network's infrastructure. However, the scale of today's Internet coupled with its distributed and heterogeneous nature makes it very challenging to acquire a complete and accurate snapshot of the topology. The purpose of this survey is to examine the main research studies that have been conducted on topics related to Internet topology discovery in the last 15-20 years and present some of the main lessons learned from these past efforts. To this end, we classify these prior studies according to the “resolution” or “level” of the topology; that is, interface-level, router-level, PoP-level and AS-level. For each resolution, we describe the main techniques and tools used for data collection, identify their major limitations and issues, and discuss the key implications that these limitations have on the quality of the collected data. In the process, we present the latest efforts in modeling the Internet's topology at the different levels and report on the role that geographic characteristics play in this context. We present the lessons learned as a checklist that every researcher working on Internet topology discovery-related problems should consult to minimize the risk of repeating some of the same or similar mistakes that have been made in the past and as a result have hampered progress in this important area of Internet research.