Determining the Location of Autonomous System Relations Among Top Tier Internet Service Provides in the United States
MetadataShow full item record
The Internet is a network of networks. Understanding where and how these networks interconnect is important for the purpose of meaningfully investigating a wide range of critical Internet-related problems, such as the vulnerability of the Internet to physical damage. While there is published work on Internet topology and structure, those studies focus primarily on finding the existence of network interconnections and characterizing the structure of the Internet by those relationships. What this thesis seeks to investigate is a methodology that helps to determine where these interconnections, or 'cross connects', are happening at the city-level. We evaluate a method for collecting cross connect data that uses geo-located vantage points and targets. We then investigate the feasibility of using distributions of round trip times to estimate a threshold for separating cross connects that occurred within a city from those that occurred outside of a city. We utilized this method to investigate the cross connects in 17 cities across the United States. Preliminary findings indicate that this method is viable in certain cities. The study also identified several trends that warrant further investigation.