Bart Rousseau

The Private User Network Interface-a GMPLS-compliant signaling interface for overlay-based multilayer networks

Feature Issue on Next-Generation WDM Network Design and Routing (WDMN). To address the integration of [IP/MPLS (multiprotocol label-switching)] packet and (optical) transport networks, the Optical Internetworking Forum (OIF) has defined a user-to-network interface (UNI) for overlay networks: the OIF UNI version 1.0, also referred to as the Public UNI. However, since its implementation agreement specification, it has become clear that this UNI--based on OIF-specific extensions to the generalized MPLS (GMPLS) protocols--will not be able to meet the future requirements of some operational environments. Hence a flexible, extensible, and fully GMPLS-compliant UNI is being defined at the Internet Engineering Task Force (IETF): the Private UNI (aka GMPLS for overlay networks). The Private UNI enhances the interaction between the packet and transport network control planes of a multilayer network architecture. This interface is particularly suitable for a control-plane interconnection model in which the equipment belongs to a common administrative entity (i.e., a single carrier environment). Signaling sessions through the Private UNI are used to invoke two main services that require constructive coordination between the network layers: provisioning and network recovery. Here we present the Private UNI specific GMPLS resource reservation protocol-traffic engineering (RSVP-TE) extensions in support of these services in IP/MPLS over SONET/SDH (synchronous optical network/synchronous digital hierarchy) GMPLS multilayer networks and illustrate how Private UNI signaling enhances coordination in multilayer network recovery.

Multi-Layer Recovery Enabled with End-to-End Signaling

Within GMPLS framework, the signaling protocol Resource reSerVation Protocol with Traffic Engineering extensions (RSVP-TE) is extended to support the requirements of an Automated Switched Optical Network architecture. This paper presents the extensions of the end-to-end connection services in an overlay network built on two control planes. RSVP-TE protocol extensions are first described between an IP/MPLS router and a SDH/GMPLS core optical cross-connect, defining GMPLS-UNI. Dimensioning of three scenarios proving the benefits of GMPLS-UNI is discussed.

Application of the Link Management Protocol to Discovery and Forwarding Adjacencies

The link management protocol (LMP) is used to maintain control channel connectivity, to verify the physical connectivity of data channels, to correlate link property information at both endpoints of a data link and to aggregate multiple data links into a single traffic engineering (TE) link. LMP also provides assistance in fault localization, both in opaque and transparent optical networks. In this paper, we show how LMP relates to the generalized multi-protocol label switching (GMPLS) suite of protocols and integrates into a distributed GMPLS control plane. We describe the role of LMP in two applications: data link (auto-)discovery and the establishment of forwarding adjacency label-switched paths (FA-LSPs) including their initiation, verification and bundling, and conclude that they can be built upon basic functions of LMP, namely control channel management, link verification and link property correlation.

Dynamic weights for OCSPF route computation in optical networks

In current routing algorithms for wavelength switched optical transport networks, based on the Open Constrained Shortest Path First (OCSPF) concept, the effects of traffic dynamicity (the fact that lightpaths (a.k.a. optical connections or optical LSPs) are continuously being set-up and torn down) are not taken into account. Therefore, in this study four new link metrics are proposed, each of which could be used to replace the current metrics being used. Two metrics are based on the average load ρ on a link during a given timeframe. Two other metrics are based on an estimate of the number of setup requests nbr that will be blocked during the lifetime of the lightpath that is being set up. It is shown that the former two link metrics (i.e. those based on ρ) significantly reduce the blocking probability compared to the case in which a constant link metric is used. The latter two metrics (i.e. those based on nbr) however perform only slightly better or even worse than the original algorithm. In addition, it is observed that by taking into account the traffic dynamicity using the average load on a link, the amount of sustainable demand can be increased by about 10% for a given blocking probability.

Capacity Efficiency of Distributed Path Restoration Mechanisms in Optical Mesh Networks

In this study the restoration performance of two closely related Sender/Chooser-based distributed path restoration protocols (both extensions of the Self-Healing Network (SHN) to path restoration) is compared. Some pathological situations in which these Sender/Chooser-based restoration algorithms perform suboptimally are identified and, where available, possible solutions are proposed. Built-in mechanisms to resolve contention between Sender/Chooser pairs were found to be very helpful. In addition, the performance of the distributed restoration algorithms studied was found to be close to its theoretical upper bound, suggesting that the pathological situations described may not be that important for real-life networks.