Optimization of IP Networks in Various Hybrid IGP/MPLS Routing Schemes.
Eueung Mulyana, Ulrich Killat · 2004
This paper presents an offline traffic engineering (TE) approach in several hybrid IGP/MPLS environments. Though IGP (Interior Gateway Protocol) routing has proven its scalability and reliability, effective traffic engineering has been difficult to achieve in public IP networks because of the limited functional capabilities of conventional IP technologies. MPLS (Multi-Protocol Label Switching) on one hand enhances the possibility to engineer traffic on IP networks by allowing explicit routes, but on the other hand it suffers from the scalability problem. Hybrid IGP/MPLS approaches rely on IP native routing as much as possible and use MPLS only if necessary. In this work we propose a novel hybrid traffic engineering method based on genetic algorithms and investigate it in several IGP/MPLS routing schemes. This can be considered as an offline TE approach to handle long or medium-term traffic variations. In our approach the maximum number of hops as well as the maximum delay an LSP (Label Switched Path) may take and the maximum number of LSPs that can be installed in the network are treated as constraints. We apply our method to the German scientific network (G-WiN) for which a traffic matrix was randomly generated. We will show results for several hybrid routing schemes and compare them to the result of pure IGP routing.