An approach to optimal peering between autonomous systems in the Internet

This paper describes a framework that addresses the important problem of optimal location of peer points for data exchange between autonomous systems in the Internet. The decision objective is to minimize the cost of peering, and at the same time improve the efficiency of inter-domain traffic flows. A combinatorial optimization formulation is used to model this problem, and location theory is applied to construct an integer zero-one mathematical program. Solution methodologies for the resulting mathematical program are surveyed and practical implementation considerations are discussed.