Multi-Perspective Optimization of GEANTInter-domain Traffic

Interdomain traffic engineering is a key issue when resource optimization across autonomous systems boundaries is concerned. Mapping interdomain traffic flows into existing service level agreements is, in general, a complex problem, for which an optimization tool is needed. Th is work aims at demonstrating the advantages of using such a tool to perform offline interdomain traffic optimization, using a multiperspective 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.