Stalled information based routing in multi-domain multilayer networks

In multilayer (e.g. GMPLS/ASTN) networks the routing becomes very complex since it is performed over multiple network layers simultaneously. Furthermore, to achieve a certain level of QoS not only topology but also link-state information has to be spread. To obtain a path from a source to a desired destination in short time, source routing is preferred that relays upon link-state information that was flooded in advance. The often the flooding is performed the more up-to-date the state information is. Unfortunately, in case of larger networks this results in a heavy signaling load (overhead). To make such a routing scheme scalable either the flooding has to be rarefied or the network has to be split hierarchically into domains and appropriate intra-domain and inter-domain routing schemes have to be applied. In this paper we propose a network model that performs topology aggregation to reduce the amount of information flooded as well as a routing scheme for these multi-layer multi-domain networks that works over that "simplified" network. Then we investigate the impact of stalled information (rare flooding) and impact of routing re-trials onto the blocking ratio. The proposed framework is a viable solution for the control plane of future scalable transport networks like GMPLS/ASTN.