Multi-perspective optimization of GEANT inter-domain traffic

Inter-domain traffic engineering is a key issue when resource optimization across autonomous systems boundaries is concerned. Mapping inter-domain traffic flows into existing service level agreements is, in general, a complex problem, for which an optimization tool is needed. This work aims at demonstrating the advantages of using such a tool to perform off-line inter-domain traffic optimization, using a multi-perspective approach. This optimization approach was applied to GEANT, the European Research and Educational Network. The paper presents optimization solutions from various perspectives, namely bandwidth, monetary cost and routing trustworthiness. Results show the costs of the original GEANT traffic engineering solution and the costs of the optimized solutions side-by-side. This case study also shows the advantages of being able to select between the perspectives that best fit the domain management policies.