Offline Scheduling Schemes to Transfer Voluminous Deadline Complying Data in Elastic Optical Networks

Abstract

For serving traffic in the inter data centers which provide services such as, duplication of data and migration of the virtual machines, it is requisited to transfer voluminous data for which, under guarantee of a finishing time within the stipulated (i.e., pre-set) deadline, specific latency is tolerable. In the current work, we propose offline routing and spectrum assignment (RSA) schemes in view of the transfer of deadline complying voluminous data demands in elastic optical networks. The proposed schemes, which jointly optimize time and frequency domains, are initially formulated as an integer linear program (ILP). Subsequently, in view of practicality, we propose scalable scheduling techniques which combine three methods of ordering demands, and two schemes of RSA. To evaluate the proposed ILP model and the scheduling methods, we conduct simulations considering realistic network parameters and topologies. The obtained results demonstrate that in comparison to ILP model, scalable methods obtain similar spectrum usage performance within reasonable times. Lastly, based on the results, we also provide a ‘rule-of-thumb’ on the selection of appropriate scheduling technique.