{"title":"Inferring the average as path length of the Internet","authors":"Cun Wang, Zhengmin Li, Xiaohong Huang, Pei Zhang","doi":"10.1109/ICNIDC.2016.7974603","DOIUrl":null,"url":null,"abstract":"The average path length of the Internet autonomous system (AS) level topology was decreasing slightly in the last decade. Researches had shown that the average AS path length of the Internet was about 3.9 hops until 2012. However this inference was based on BGP routing tables collected from some vantage points that cannot cover the whole Internet. Researchers had also developed different methods to infer AS paths of the Internet, the easiest method considered the shortest paths as AS paths, but the reality is the ASes have self-defined routing policies which lead to a longer path than the shortest path. Other methods improved the inference accuracy. But they were also time consuming and difficult to be used in large scale. To explore the average AS path length of the Internet, we developed a new method to infer AS paths in a comparatively complete AS level topology. The core of this method was a path decision algorithm that simulated the packet transiting process for each AS in the Internet. The inference paths generated from this method were called as decision paths. The experimental results indicated that the decision paths were close to the actual AS paths especially for the path length distribution. At last, we discussed the effect of the incompleteness of peering links in the topology and inferred the average AS path length of the whole Internet using decision paths.","PeriodicalId":439987,"journal":{"name":"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICNIDC.2016.7974603","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9
Abstract
The average path length of the Internet autonomous system (AS) level topology was decreasing slightly in the last decade. Researches had shown that the average AS path length of the Internet was about 3.9 hops until 2012. However this inference was based on BGP routing tables collected from some vantage points that cannot cover the whole Internet. Researchers had also developed different methods to infer AS paths of the Internet, the easiest method considered the shortest paths as AS paths, but the reality is the ASes have self-defined routing policies which lead to a longer path than the shortest path. Other methods improved the inference accuracy. But they were also time consuming and difficult to be used in large scale. To explore the average AS path length of the Internet, we developed a new method to infer AS paths in a comparatively complete AS level topology. The core of this method was a path decision algorithm that simulated the packet transiting process for each AS in the Internet. The inference paths generated from this method were called as decision paths. The experimental results indicated that the decision paths were close to the actual AS paths especially for the path length distribution. At last, we discussed the effect of the incompleteness of peering links in the topology and inferred the average AS path length of the whole Internet using decision paths.