Rearrangeable multihop lightwave networks: congestion minimization on regular topologies

Abstract

We address the design of rearrangeable multihop lightwave networks having regular connectivity topology (such as Perfect Shuffle (PS), de Bruijn (dB) graph, GEneralized shuffle?exchange Multihop NETwork (GEMNET) and Manhattan Street Network (MSN)). We propose a new formulation of the combined station assignment/flow routing problem with the congestion minimization objective. The formulation is applicable to any regular topology. We then develop a heuristic solution strategy based on tabu search. Given our objective, we want to assess the merit of a regular versus an arbitrary topology, as well as compare different regular topologies across several data instances. In terms of achieving small congestion (as obtained heuristically), the results suggest that with increased problem sizes, regular topologies become more attractive. In such cases the benefit of having less restricted arbitrary network topology might not be fully utilized. Equivalently, having regular topology for larger instances will still provide a large enough search space to (heuristically) obtain good throughput (i.e., small congestion). Moreover, such design will implicitly offer benefits associated with management and operations of regular topologies.