Evolution of the Internet AS-Level Ecosystem

Srinivas Shakkottai
Texas A&M University
College Station, USA

Marina Fomenkov
Cooperative Association for Internet Data Analysis - CAIDA
San Diego Supercomputer Center,
University of California, San Diego

Ryan Koga
Cooperative Association for Internet Data Analysis - CAIDA
San Diego Supercomputer Center,
University of California, San Diego

Dmitri Krioukov
Cooperative Association for Internet Data Analysis - CAIDA
San Diego Supercomputer Center,
University of California, San Diego

kc claffy
Cooperative Association for Internet Data Analysis - CAIDA
San Diego Supercomputer Center,
University of California, San Diego

Abstract. We present an analytically tractable model of Internet evolution at the level of Autonomous Systems (ASs). We call our model the multiclass preferential attachment (MPA) model. As its name suggests, it is based on preferential attachment. All of its parameters are measurable from available Internet topology data. Given the estimated values of these parameters, our analytic results predict a defnitive set of statistics characterizing the AS topology structure. These statistics are not part of model formulation. The MPA model thus closes the \measure-modelvalidate-predict" loop, and provides further evidence that preferential attachment is the main driving force behind Internet evolution.

Key words: Preferential attachment, Internet evolution, AS-level topology, Internet measurement

Download full paper : pdf