Radial Structure of the Internet

TL;DR

This paper investigates the radial structure of the Internet at the Autonomous System level, revealing a distinct core-periphery organization and highlighting the complexity of peripheral regions beyond existing models.

Contribution

It introduces new methods for analyzing AS network data from a radial perspective and compares real data with generative models to identify structural differences.

Findings

Clear core-periphery distinction in AS networks

Periphery exhibits more complex structure than models predict

New visualization methods for network analysis

Abstract

The structure of the Internet at the Autonomous System (AS) level has been studied by both the Physics and Computer Science communities. We extend this work to include features of the core and the periphery, taking a radial perspective on AS network structure. New methods for plotting AS data are described, and they are used to analyze data sets that have been extended to contain edges missing from earlier collections. In particular, the average distance from one vertex to the rest of the network is used as the baseline metric for investigating radial structure. Common vertex-specific quantities are plotted against this metric to reveal distinctive characteristics of central and peripheral vertices. Two data sets are analyzed using these measures as well as two common generative models (Barabasi-Albert and Inet). We find a clear distinction between the highly connected core and a sparse periphery. We also find that the periphery has a more complex structure than that predicted by degree distribution or the two generative models.