Joint Optimization for Multicast Provisioning in Mixed-Line-Rate Optical Networks With a Column Generation Approach

Abstract

Mixed-line-rate (MLR) optical networks (e.g., 10/40/100 Gb/s) offer a new paradigm in backbone networks to meet the increasing traffic demands. They are considered as a transitional solution for future flexible and dynamic optical networks. In order to provision multicast in MLR optical networks, we need to address multicast routing, line rate selection, and wavelength assignment subproblems, while still respecting the maximum transmission reach. In this paper, we investigate the multicast provisioning problem, which aims at minimizing the total cost on transponder, wavelength channel, and wavelength usage. First, a lightpath-based joint integer linear programming (ILP) model is formulated to minimize the total cost. To overcome the scalability of this joint ILP model, a decomposition approach based on column generation is proposed, which achieves comparable results within an extremely short period of time. Extensive simulations are performed to study the impact on the total cost of multicast provisioning with the MLR regarding the number of multicast sessions, destinations, traffic amount, and network topologies. It is observed that the multicast provisioning with the MLR has significant total cost savings over the single line rate, especially in the scenarios of a large number of multicast sessions and in optical networks with small average nodal degree.