Achim Autenrieth

Optimal configuration of p-cycles in WDM networks

We investigate the deployment of p-cycles (preconfigured protection cycles) in WDM mesh networks with and without wavelength conversion. We develop optimization models for the configuration of the cycles and apply these on a case study for a pan-European network. The results in particular for wavelength converting networks show that p-cycles achieve high efficiency.

Recovery time analysis of differentiated resilience in MPLS

This paper presents a recovery time analysis of differentiated resilience concepts in MPLS networks. In analogy to the differentiated services QoS model a differentiated resilience concept offers several levels (or classes) of resilience to IP services. The different levels of resilience can be achieved by using different MPLS recovery models like protection switching with local or global repair or on-demand restoration. Depending on the recovery model and the network scenarios, the recovery time of these resilience mechanisms may vary. After a short introduction to MPLS recovery mechanisms the and to the differentiated resilience concept this paper presents a detailed timing model for MPLS recovery mechanisms. This timing model is used in a case study to evaluate the recovery time of the resilience differentiated QoS architecture, which is using an extended QoS signaling for differentiated resilience provisioning. Finally, the recovery time results are analyzed taking into consideration recovery capacity requirements presented in previous publications.

NGL01-6: Optimal Routing and Grooming for Multilayer Networks with Transponders and Muxponders

The parallel existence of different transport technologies in todays networks is a challenging aspect for network planning and operation. Due to the grooming and multiplexing possibilities at multiple layers, the routing complexity increases dramatically. We develop a network model and an integer linear program (ILP) formulation to compute the optimal network design. The novel ldquopath over pathrdquo concept maps traffic demands into grooming flows that are provisioned as transparent lightpaths. This approach enables scalability of the problem size by limiting the number of eligible grooming configurations and transparent lightpaths. The optimization objective is to minimize the overall capital expenditures with respect to the deployed network equipment. Case study results show that the cost relation transponders/muxponders and restrictions on the equipment have a significant impact on the routing and network configuration.

RD-QoS - the integrated provisioning of resilience and QoS in MPLS-based networks

The provisioning of QoS and resilience is becoming a key issue in todays IP-based networks. The current discussion centers around multiprotocol label switching (MPLS) as the common platform to support both, QoS and resilience in IP backbones. Any well-designed recovery strategy has to take into account the different resilience requirements of the single traffic flows in order to avoid excessive usage of bandwidth for standby links. After reviewing MPLS components and recovery options, this paper presents a signaling method which allows the resilience requirements of individual services to be signaled to the network in addition to their QoS requirements. The resulting possibility to tailor the end-to-end network resilience is then used to integrate the capacity provisioning for the single resilience classes with the MPLS traffic-engineering process. A case study shows the large network capacity savings achievable via this approach.

Dual failure protection in multilayer networks based on overlay or augmented model

In this paper we present and compare different multilayer protection mechanisms which enable the client to protect its connections against dual failures in the server layer. In multilayer networks based on overlay or augmented model, the client is not aware of the routing information of the server layer. Therefore, some resilience mechanisms need new functions that have to be provided by the User Network Interface (UNI). Based on generic building blocks for surviving dual failures we propose four multilayer protection models. We extend our existing simulation software for analyzing resilience options and evaluate the selected operation scenarios.

Business Models for Next Generation Transport Networks

Control plane technologies will play a major role in next generation transport networks. Many publications and standards claim that approaches like Generalized Multiprotocol Label Switching (GMPLS) and Automatically Switched Optical Networks/Transport Networks (ASON/ASTN) will simplify network operation and will enable new services. In this paper, an approach is presented to analyze the business processes of transport network operators. The main operator processes are mapped to the well-known value chain concepts of Porter. This allows investigating the advantages of control plane technologies from an economic point of view. This paper shows that the introduction of control plane technology affects the business of the individual operators as well as the overall value system they form as a virtual enterprise. Moreover new business opportunities are identified and two business cases are presented.