Section: New Results

Algorithms: Bandwidth Allocation in Optical Networks

Participants : Christine Fricker, Philippe Robert, James Roberts.

The development of dynamic optical switching is widely recognized as an essential requirement to meet anticipated growth in Internet traffic. Since September 2009, RAP has investigated the traffic management and performance evaluation issues that are particular to this technology. A first analysis of passive optical networks used for high speed Internet access led to the proposal of an original dynamic bandwidth allocation algorithm and to an evaluation of its traffic capacity. Our activity on optical networking is carried out in collaboration with Orange Labs with whom we have a research contract. We have also established contacts with Alcatel-Lucent Bell Labs and had fruitful exchanges with Iraj Saniee and his team on their proposed time-domain wavelength interleaved networking architecture (TWIN).

We have analyzed the traffic capacity of wavelength division multiplexing (WDM), passive optical networks (PONs) where user stations (optical network units) are equipped with tunable transmitters. For these systems users can use any of the multiple wavelengths to transmit their data but only within the limit determined by the number of transmitters they possess. A mean field approximation is used to estimate the capacity of a limited-gated multiserver polling system with a limit on the number of servers a given station can use simultaneously. The approximation provides an expression for the stability limit under very general assumptions about the traffic process and system configuration.

In 2011, we began work on bandwidth allocation in meshed networks. We have evaluated the TWIN architecture in a metropolitan area network with an original medium access control (MAC) algorithm. This algorithm was inspired by our prior work on access networks and ensures an efficient and fair allocation of bandwidth to flows between network nodes.

The TWIN architecture is not extensible to a wide area for reasons of scalability and the excessive signalling delay between geographically distant nodes. We have therefore invented a new notion of a multipoint-to-multipoint lightpath that avoids these problems. A patent relating to this invention has been granted. This patent is owned by Orange following the terms of our contract with them. The paper [16] describes the invention and its evaluation. A major advantage demonstrated in this paper is the energy saving achieved by the use of the proposed optical technology in place of electronic routers. An extended version of the paper has been accepted for publication in Journal of Optical Communication and Networking [24] .

Ongoing research seeks to apply this type of networking solution to data centres, on one hand, and to geographically spread tier-1 Internet carrier networks, on the other. Some of this work is performed in collaboration with Orange Labs under the terms of our research contract. An interesting new development is the application of new coherent optical technology that allows tunable receivers as well as tunable transmitters. We are evaluating the performance of a bandwidth allocation algorithm that exploits this technology.

A wider reaching collaboration has been established under the terms of a Celtic Plus project called SASER. This project was approved by the EU in 2012 and funding has been obtained for our participation from the French authorities. The project kickoff meeting was held in November 2012. Our contribution relates to the use of TWIN to create an extended metropolitan optical network. Our partners in the corresponding work package task are Orange, Telecom Bretagne and the engineering school ENSSAT. Overall reponsibility for the work package (where alternative optical network architectures are also evaluated) is with Alcatel-Lucent Bell Labs.