ACM Home Page
Please provide us with feedback. Feedback
Public review for the internet AS-level topology: three data sources and one definitive metric
Full text PdfPdf (17 KB)
Source ACM SIGCOMM Computer Communication Review archive
Volume 36 ,  Issue 1  (January 2006) table of contents
FEATURE: Reviewed articles table of contents
Pages: 15 - 16  
Year of Publication: 2006
ISSN:0146-4833
Author
Michalis Faloutsos  University of California Riverside, USA
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 2,   Downloads (12 Months): 16,   Citation Count: 0
Additional Information:

abstract   index terms   collaborative colleagues  

Tools and Actions: Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1111322.1111327
What is a DOI?

ABSTRACT

The paper provides an overview of the current available data sets for AS level topology and examines the usefulness and intuition of many and most widely used metrics used for graph analysis, and highlight some less known metrics. The paper makes two practical contributions.First, the paper highlights an important issue: all source of AS information are not the same. The paper utilizes three sources of information BGP tables, traceroute-based and archived information (WHOIS/IRR). The observed topo-logical properties are fairly different. A potential explanation could be that none of the sources captures the whole AS graph. Thus, the measurement community has more work to do to obtain a complete topology.Second, the paper critically analyzes the importance of the different topology metrics. The last six years have seen a proliferation of new metrics that are used in a selective fashion from researchers. This is one of the first thorough studies of most known metrics and on top of that many metrics that are used in related fields that also study networks (e.g. physics).In conclusion, this paper could become a reference point for making the analysis of the AS topology more systematic both in terms of making researchers: (a) aware of the differences among the datasets, and (b) careful in selecting metrics.