1: \documentclass{sigcomm-alternate}
2: \usepackage{graphicx,amsmath,amssymb,url,subfigure}
3: \begin{document}
4:
5: \title{The Internet AS-Level Topology: \\ Three Data Sources and One Definitive Metric}
6:
7: \numberofauthors{1}
8:
9: \author{
10: Priya Mahadevan \\ UCSD \and
11: Dmitri Krioukov \\ CAIDA \and
12: Marina Fomenkov \\ CAIDA \and
13: Bradley Huffaker \\ CAIDA \and
14: Xenofontas Dimitropoulos \\ Georgia Tech \and
15: kc claffy \\ CAIDA \and
16: Amin Vahdat \\ UCSD \and
17: {\normalsize \{pmahadevan,vahdat\}@cs.ucsd.edu, \{dima,marina,brad,fontas,kc\}@caida.org}
18: }
19:
20: \maketitle
21:
22: \begin{abstract}
23: \noindent
24: We calculate an extensive set of characteristics for Internet AS
25: topologies extracted from the three data sources most frequently used
26: by the research community: traceroutes, BGP, and WHOIS. We discover that
27: traceroute and BGP topologies are similar to one another but differ
28: substantially from the WHOIS topology. Among the widely considered
29: metrics, we find that the {\em joint degree distribution} appears to
30: fundamentally characterize Internet AS topologies as well as narrowly
31: define values for other important metrics. We discuss the interplay
32: between the specifics of the three
33: data collection mechanisms and the resulting topology views. In particular, we show
34: how the data collection peculiarities explain differences in the
35: resulting joint degree distributions of the respective topologies.
36: Finally, we release to the community the input topology
37: datasets, along with the scripts and output of our calculations. This
38: supplement should enable researchers to validate their models against
39: real data and to make more informed selection of topology data sources
40: for their specific needs.
41: \end{abstract}
42:
43: \category{C.2.5}{Local and Wide-Area Networks}{Internet}
44: \category{C.2.1}{Network Architecture and Design}{Network topology}
45: \category{G.3}{Probability and Statistics}{Distribution functions,
46: multivariate statistics, correlation and regression analysis}
47: \category{G.2.2}{Graph Theory}{Network problems}
48:
49: \terms{Measurement, Design, Theory}
50:
51: \keywords{Internet topology}
52:
53: \section{Introduction}
54: \label{sec:intro}
55: \input{intro}
56:
57: \section{Data Sources}
58: \label{sec:data}
59: \input{data}
60:
61: \section{Topology Characteristics}
62: \label{sec:characteristics}
63: \input{characteristics}
64:
65: \section{Conclusion}
66: \label{sec:conclusion}
67: \input{conclusion}
68:
69: \section{Acknowledgments}
70: \input{ack}
71:
72: \input{paper.bbl}
73:
74: \end{document}
75: