A Geographic Directed Preferential Internet Topology Model

The goal of this work is to model the peering arrangements between Autonomous Systems (ASes). Most existing models of the AS-graph assume an undirected graph. However, peering arrangements are mostly asymmetric Customer-Provider arrangements, which are better modeled as directed edges. Furthermore, it is well known that the AS-graph, and in particular its clustering structure, is influenced by geography. We introduce a new model that describes the AS-graph as a directed graph, with an edge going from the customer to the provider, but also models symmetric peer-to-peer arrangements, and takes geography into account. We are able to mathematically analyze its power-law exponent and number of leaves. Beyond the analysis we have implemented our model as a synthetic network generator we call GdTang. Experimentation with GdTang shows that the networks it produces are more realistic than those generated by other network generators, in terms of its power-law exponent, fractions of customer-provider and symmetric peering arrangements, and the size of its dense core. We believe that our model is the first to manifest realistic regional dense cores that have a clear geographic flavor. Our synthetic networks also exhibit path inflation effects that are similar to those observed in the real AS graph.

Abstract licence: CC0

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.