Optimal Algorithms for Near-Hitless Network Restoration via Diversity Coding

Abstract: Diversity coding is a network restoration technique which offers near-hitless restoration, while other state-of-the art techniques are significantly slower. Furthermore, the extra spare capacity requirement of diversity coding is competitive with the others. Previously, we developed heuristic algorithms to employ diversity coding structures in networks with arbitrary topology. This paper presents two algorithms to solve the network design problems using diversity coding in an optimal manner. The first technique pre-provisions static traffic whereas the second technique carries out the dynamic provisioning of the traffic on-demand. In both cases, diversity coding results in smaller restoration time, simpler synchronization, and much reduced signaling complexity than the existing techniques in the literature. A Mixed Integer Programming (MIP) formulation and an algorithm based on Integer Linear Programming (ILP) are developed for pre-provisioning and dynamic provisioning, respectively. Simulation results indicate that diversity coding has significantly higher restoration speed than Shared Path Protection (SPP) and p-cycle techniques. It requires more extra capacity than the p-cycle technique and SPP. However, the increase in the total capacity is negligible compared to the increase in the restoration speed.