Virtualisation of optical networks utilising squeezing protection: the exact formulation

Abstract

Elastic optical network (EON) is a new technology used for virtualisation of optical networks. In this Letter, a new linear formulation with protection is proposed for the design of EON networks with several virtual optical networks (VONs). The mixed integer linear programming (MILP) formulation integrates the concepts of bandwidth squeezing, grooming and modulation format, and allows different protection capability in each VON. The proposed MILP provides efficient survivability results and resource savings (in terms of the spectrum) for a complete design of EON with virtualisation.

Virtualization of elastic optical networks and regenerators with traffic grooming

“…This virtualization will enable the rapid deployment of multi-tenant, application-specific and customized virtual infrastructures (VIs). In the optical network, each VON can have its own control plane to enable the provisioning of dynamic, adaptive and fault-tolerant optical network services [15].…”

“…Notice that the slots of a virtual link that passes through a node in the substrate network have to be entirely switched to a single output link. Constraint (15), along with the Objective Function, is used to minimize the maximum number of spectrum slots used among all demands. Constraint ( 16) is responsible for setting the value of A ij,t mn , which is set to one whenever traffic from the virtual link i − j in t is routed through the physical link mn.…”

“…This paper examines the emerging role of virtualization in the context of some important aspects, specifically: 1) a comparison between the grooming and non-grooming approaches, showing the advantages of grooming; 2) the role of virtual elastic regenerators (VERs), [7], in virtualization design as a solution for efficiently regenerating various bandwidth superchannels (because they are cost-efficient in relation to traditional dedicated data-rate regenerators or rate-adaptive regenerator [13]); 3) proposing heuristics to guide the solution for large instances as this type of problem is non-deterministic polynomial time (NP) complete [14]; 4) demonstrating MILPs as benchmarks for the design problem of VONs with other objectives considering elastic regenerators in constraints, for example, a survivability study, such as that presented in [12,15,16].…”