Predicting and choosing alternatives of route updates per QoS VNF in SDN

In this work, we model route updates as part of Quality of Service (QoS) Virtual Network Function (VNF) in Software-Defined Networking (SDN) architecture. In SDN, the control plane and the data plane are decoupled. This allows high flexibility by providing abstractions for network management applications and being directly programmable. However, reconfiguration and updates of a network are sometimes inevitable due to policy changes, maintenance or failures. The update process should not form congestion in the network, nor result in time delays, and not lose any packets. It may be the case that there exist several alternatives for a route update. In the extreme case, there is a chance that one of the possible existing routes is much longer, however, a more reliable one, and as such will update faster. Taking into account the random nature of the network functioning, we supplement the proposed algorithm by Delaet et al. for route updates with a technique based on Markov chains. In our extension of the algorithm, we describe the transmission of packets through a chosen route for comparing the update process for all the possible alternative routes. The virtualized service implementing the update algorithm will provide a recommendation for an optimal choice of a route based on the performed calculations. Fundamentally, we create a QoS VNF for seamlessly updating a route, regarding network parameters, and taking into consideration the complexity and possible failures of updating a route.