AStop: A New Topology Generator at the Autonomous Systems Level

Abstract

Considering the degree of an autonomous systems (ASs) as the number of edges or links of an AS, several studies have shown that the distribution of the degree of an AS follows power-laws and that its cluster coefficient is higher than that found in random networks. We introduce a new model which is based on i) observations obtained from the analysis of the union of six BGP tables available publicly, and ii) the business relationships between ASs. We compared the performance of our model with models widely cited and two real ASs topologies. Our model performs well for the following parameters: number of ASs with degree equal to one, the AS with the maximum degree achieved and CCDF