The article tackles the problem of quality of service assurance in photonic networks. The idea of multi-servicephotonic network model with the coexistence of optical circuit and packet switching mechanisms and cell communication is used as a basis for service differentiation in the optical domain. Cell loss ratio as a key performance indicator determines the required optical switching mechanism. Service provisioning is performed using call admission control mechanism with real-time cell loss ratio estimation procedure. Service blocking probability calculation utilizes discrete event simulation of service provision and teardown requests applied to core network topology from COST 266 project. Three simulation scenarios are included in the analysis – pure optical packet switching network, and coexistence of optical packet and circuit switching with and without possibility of communication redirection between the switching mechanisms. Simulation scenarios are additionally altered with the cell loss ratio constraint and number of delay lines.

The article tackles the problem of optimization methods in optical network design process, based on optimal traffic routing with the goal to minimize the utilized network resources for given topology and traffic demands. An optimization framework Nyx has been developed with the focus on flexibility in solving optimization problems by implementing general heuristic search techniques. Nyx modular organization has been described, including coding types for solutions and genetic algorithm as the optimization method. Optimal routing has been implemented to demonstrate the use of Nyx in the optical network design process. Optimal routing procedure has been applied to Pan-European optical network with variations of routing procedures and the number of wavelengths. The analysis included no protection scenario, 1+1 protection and path restoration. The routing was performed using shortest path routing and optimal routing which minimizes the use of optical network resources, being network multiplexers, amplifiers and fibers.

Approaches to Performance Analysis of Packed Switched Optical Ring An optical ring network employing optical packet switching is investigated. An analytical approach allowing the computation of end-to-end cell loss ratio (CLR) is proposed. The CLR dependence on topological parameters (number of stations), network parameters (number of wavelengths), and node structure (buffer capacity) is computed assuming a uniform traffic load. A discrete event simulation confirms the analytical results.

Network resilience is an issue of deep concern to network operators being eager to deploy high-capacity fibre networks, since a single failure in the network could result in significant losses of revenue. The importance of network reliability will keep pace with the steadily increasing network capacity. For very-high-capacity future optical networks, carrying multitudes of 10 Gbit/s channels per fibre strand, a failure of optical connection will interrupt a vast amount of services running on-line, making the connection availability a factor of great significance. Therefore the ultrahigh capacity future optical networks will face a challenge of providing very efficient and fast survivability mechanisms. In this chapter we review the terminology and basic resilience techniques along with the results of research work on optical network survivability performed in the frame of COST291 cooperation. Our research work was focused on reliability performance improvement and on recovery in multilayer optical networks.

In this paper, the impact of dependent failures of optical links on the availability of the optical network is analyzed using Monte Carlo simulation. Simulation is used because analytical approach to availability calculation is too complex when there are dependencies between failures. Availability model is described and used in two case studies: ring topology and several functional structures of two wavelength paths. Results of the simulation for the case studies are analyzed and compared. The results of the simulation are verified using analytical availability calculations for a simple structure without dependent failures.

U ovom radu opisano je modeliranje i simulacija zavisnih kvarova optickih linkova u optickoj mreži. Za proracun raspoloživosti koristena je Monte Carlo simulacija zbog prevelike složenosti analitickog proracuna. Utjecaj zavisnosti između kvarova analiziran je na pokaznom primjeru. Rezultati pokazuju da zavisnost između kvarova može znacajno utjecati na raspoloživost opticke mreže i da se zavisnost između kvarova ne bi trebala zanemarivati u proracunima raspoloživosti.

The article describes the philosophy of component approach to system design, illustrated on the optical transmission network example. The approach is based on the object oriented paradigms of Cosmos and Nyx tools. Cosmos serves as the support for system and component development, behavior description, and simulation and analytic calculation development. Nyx in turn enables description of optimization procedures using general heuristic search techniques. The combination of these two development environments and their component approach enables reusability and shortening of the new application development time.

The paper provides a logical description of the link protection scheme implemented in the photonic transmission network. Description is based on the notion of transport entities defined on the link level, and extended by resilience domains. The domains are used to relate primary and protection transport entities, being the crucial information for the traffic rerouting after a failure occurs. The protection capacity reservation depends on the protection probability of the primary communication, what has been introduced as a design option. The performance of the link protection scheme has been evaluated regarding the communication unavailability and the network price expressed as fibre lengths

This paper addresses the basic concepts of the packet switching in the optical domain and describes an analytical approach to evaluate the end-to-end performance of networks employing slotted (fixed length) optical packets. Thus, for a given topology and traffic matrix, the end-to-end cell loss ratio is analytically computed assuming an uncorrelated traffic. A network dimensioning method relying on this approach is also presented.

The paper discusses issues related to the support of the p-Cycles concept in the Multi-Service Photonic Network (MSPN), with the emphasis on the design issues. Design procedure is based on general heuristic search techniques such as genetic algorithm. Performance measures have been defined to evaluate the solution quality. The design procedure has been applied to the Pan European topology, with the analysis of obtained p-cycle protection performance results.

The article gives an overview of techniques for calculating the service blocking probabilities in core photonic network, with the assumption of static routing and optical circuit switching with and without wavelength conversion. An example of algorithm without link load correlation assumption has been explained and mathematically defined. It has been applied on the Cost 266 topologies, illustrating the difference between wavelength conversion and no wavelength conversion cases. Reduced load model was an example of link load correlated model, attributed with the detailed construction explanation. Results of the uncorrelated, correlated and simulation blocking probability calculation have been compared on a network example.

The article proposes a classification of resilience schemes in photonic transmission network and provides examples of protection and restoration schemes applied in mesh networks. The overview of resilience performance measures has been presented. Cost 266 topologies have been compared using total fiber length, and mean and worst network unavailability results.

The paper presents the Multi-Service Photonic Network (MSPN) architecture, implementing the idea of optical circuit, burst and packet switching coexistence. We focus on the architecture using a layered perspective, and describe each layer functionality and technical implementation. Generic node structure gives a systematic description of MSPN functionalities through switching elements and transport interfaces. The implementation of new switching paradigms is particularly addressed by the discussion of the ways of transporting possibilities of optical packet/burst switching services over optical circuit switching services optical packet/burst. The MSPN support of different topological paradigms is of essential importance for the next generation optical transmission network deployment.

Keywords: optical packet switching ; network dimensioning Reference EPFL-CONF-149282 Record created on 2010-06-09, modified on 2016-08-08

The paper focuses on issues related to the modeling of photonic network availability. An example of photonic network architecture is presented together with the services it can provide to clients. The paper gives an overview of quantitative survivability measures and further concentrates on availability performance. Generic availability analysis and its elements, models and calculation procedures as well as a generic tool for availability analysis are explained.

The paper presents application of Cosmos framework in the education on communication networks. It gives a brief overview of Cosmos and main steps in application development. Of the four presented application modules three illustrate the concepts at different layers of communication networks. These are optical burst switching, gigabit Ethernet and TCP. An application that illustrates availability and reliability-related concepts is presented.

This paper gives an overview of the work carried out during the course of the COST-266 action on circuit/wavelength switched network studies. Several subjects are discussed in detail in this paper. A first important topic deals with the different control plane models in single- and multidomain networks. A second important issue is the routing and wavelength assignment problem, both in networks with static and dynamic traffic demands. Also the IP-over-OTN network architecture, envisaged to be the network architecture most suited for future transport networks, is discussed. Finally the study on optical virtual private networks is presented.

Availability is an important performance feature of a telecommunication system and network. Failures of components could cause interruption of telecommunication services. thus diminishing user satisfaction and generating losses to network operator. Basic definitions of reliability, availability and security are done, as well as related measures: failure rate, fit, repair rate, MTTF, MTBF, MTTR and MDT. Availability is analysed by bottom up approach which includes levels of components, system and network. Availability structures are considered, taking into account different options of introducing redundancy. Availability models are mostly based on Markov model of repairable systems. All relevant parameters influencing availability function are considered, especially dependency and disjointedness of failure occurrences. Availability data of optical components are listed. Availability models of photonic network are described concerning hierarchy of instances: wavelength channel, wavelength path, logical channel and logical connection. Behaving of the network instances in the case of failure are described by Boolean expressions. Usual assumptions and approximations in the models are done. Different topologies and protection & restoration scenarios in photonic network are analyzed. In availability evaluation two basic approaches are used analytical and simulation (Monte Carlo). In order to find how part of a network influence on total availability figure sensitivity analysis could be used. Examples of topologies and scenarios will he explained: point-to-point photonic communications systems, WDM systems with(out) protection, photonic rings based on WDM, WDM meshed photonic networks and p-Cycles. Tools for availability modeling and evaluation are described. TUTORIAL OUTLINE