Energy-efficient network planning and traffic provisioning in IP-over-elastic optical networks

Abstract

In this paper, for IP-over-elastic optical networks (IP-over-EONs) we explore energy-efficient network planning and traffic provisioning methods that exploit the flexibility of sliceable bandwidth variable transponders (SBVTs) with reference to providing multiple transmission options and setting up multiple independent optical channels sourced (destined) from (to) a single SBVT. From the same SBVT, several optical channels with different modulation formats, transmission spectra, capacities, maximum transparent reach limits, and related power consumption (PC) values can be set up. For both the static and dynamic traffic conditions, with due consideration of efficiency in overall PC of network (comprising all major power-consuming equipment), the most suitable optical channels out of all the candidate options are set up and traffics are provisioned along the most suitable paths. We judiciously use electrical and optical layer traffic grooming to reduce PC. We present a mixed integer linear programming (MILP) model and an auxiliary graph (AG)-based heuristic scheme for energy-efficient network planning and static traffic provisioning in IP-over-EON. We validate through simulation the result of the heuristic scheme with the conclusion drawn from the result of the MILP model. We also present an AG-based heuristic scheme for energy-efficient traffic provisioning in dynamic traffic condition and show result through exhaustive simulations for realistic network setting.