Towards a Precise and Complete Internet Topology Generator

Correctly modeling the Internet topology is critical for the design of the next-generation network protocols. In this paper we present a more rigorous evaluation of the positive-feedback preference (PFP) model by comparing the model against a measurement of the Internet topology at the autonomous systems (AS) level which was collected later than that used in the previous evaluation, and a recent measurement of the Chinese Internet AS graph which is a local subgraph evolved fairly independently from rest of the global graph. Our numerical simulation shows that the PFP model, using the same parameters, precisely reproduces both the global AS graph and the AS subgraph in China. This work increases our confidence that the PFP model is a precise and complete Internet topology generator and suggests that the model's grow mechanisms have successfully captured the fundamental dynamics that universally govern the Internet evolution