Delay aware survivable routing with network coding in Software Defined Networks

It was demonstrated in transport networks that network (diversity) coding can provide sufficient redundancy to ensure instantaneous single link failure recovery, while near-optimal bandwidth efficiency can be reached. However, in the resulting multi-path routing problem the end-to-end delays were not considered. On the other hand, even in a European-scale network the delay difference of the paths has severe effect on the Quality-of-Service of application scenarios, such as video streaming. Thus, in this paper we thoroughly investigate survivable routing in Software Defined Networks (SDNs) with several additional delay bounds to the bandwidth cost minimization problem. We build on the fact that, if the user data is split into at most two parts, then the minimum cost coding solution has a well-defined acyclic structure of subsequent paths and disjoint path-pairs between the communication end-points. Complexity analysis and integer linear programs are provided to solve these delay aware survivable routing problems in SDNs.