Master’s thesis is dedicated to the issue of network element. In the first part are described technological aspects as the internal structure of network elements and control mechanisms that provide differentiaton of services and support their quality requirements. In the first chapter are listed control mechanisms from most simple like FIFO to the more sophisticated like CBWFQ. Active queue managements are not missing and they are described in the section Riadiaci člen. Significant role in the differentiation of services have processes marking and pakets classification. The second chapter is devoted most promising standard DiffServ. Contains section devoted to architecture, paket marking, code point DSCP and the final part is devoted to the types of per-hop behavior and it is expedited forwarding EF and technique assured forwarding AF. The third chapter gives an overview common models of neural networks, their properties and assess their suitability for deployment in network elements. The routher design itself and its structure in programming environment Matlab/Simulink is the subject of the fourth chapter. Except used bloks from Simulink library is described their setting and function in the wiring. Consequently results are reviewied and conclusions drawn.

Le routage multi-chemins est une technique qui permet l'équilibrage de la charge en multiplexant les flux vers une destination sur plusieurs chemins. Nous proposons une approche de routage multi-chemins qui, peut être schématisée en deux étapes : l'étape de choix des chemins candidats et l'étape de distribution du trafic sur un sous-ensemble de ces chemins. <br />Dans ce travail, nous avons commencé par effectuer une étude comparative d'un point scalabilité et stabilité, de plusieurs algorithmes de routage multi-chemins, basés sur MPLS. Cela nous a permis de retenir WDP pour la sélection des chemins candidats et LDM pour la distribution des demandes de trafic reçues par un routeur entrant du réseau d'un FAI. Nous proposons dans ce travail PER, un algorithme qui est une amélioration de l'algorithme de distribution de LDM. Ces différents algorithmes nous ont permis de réaliser plusieurs algorithmes « hybrides » dont LBWDP (Load Balancing over Widest Disjoint Paths) qui par simulation a été prouvé comme étant un algorithme plus performant que des modèles comme LDM ou MATE. <br />Pour une meilleure garantie de QdS nous avons cherché à intégrer la différentiation de service (DiffServ) avec notre technique d'ingénierie de trafic (DiffServ-aware MPLS Traffic Engineering : DS-TE). Nous proposons PEMS (PEriodic Multi-Step algorithm for DS-TE network) comme un modèle de DS-TE pour différencier la qualité du service selon la classe du trafic. Dans ce cadre, nous proposons un nouvel algorithme de sélection de chemins candidats en fonctions des critères préférentiels de chaque classe de trafic. L'utilisation de PER permet ensuite de distribuer en fonction de critères dynamiques les demandes reçues sur les meilleurs chemins de chaque classe. Par simulation à l'aide de ns-2, nous avons montré que PEMS répartie moins bien la charge que LBWDP mais que les classes EF et AF ont une qualité de service meilleure que dans le cas de LBWDP.

The motivation for the research carried out was to develop a rule based traffic management scheme for DiffServ networks with a view to introducing QoS (Quality of Service). This required definition of rules for congestion management/control based on the type and nature of IP traffic encountered, and then constructing and storing these rules to enable future access for application and enforcement. We first developed the required rule base and then developed the software based mobile agents using the Java (RMI) application package, for accessing these rules for application and enforcement. Consequently, these mobile agents act as smart traffic managers at nodes/routers in the computer based communication network and manage congestion.

L'Internet sert de support de communication à un grand nombre d'applications dans le cadre des réseaux d'entreprise. Cependant, un certain nombre d'applications multimédia (e.g. téléphonie IP) nécessite un support de mécanisme de qualité de service dans le réseau. Chaque mécanisme du QoS est provisionné en fonction du contrat (SLA - Service Level Agreement) établit entre l'utilisateur/application et l'ISP (Internet Service Provider). Le rôle important de la gestion du QoS est de conserver toutes les caractéristiques établies par le SLA pendant toute la durée du contrat. Cette gestion peut être statique ou dynamique. Dans le cas statique, la gestion de la QoS dans le réseau est effectuée d'une manière expérimentale et non-efficace, manuellement par l'administrateur de réseau sur une grande échelle de temps. La gestion dynamique de la QoS est effectuée en utilisant des mécanismes adaptables et de façon automatique pour une gestion flexible et efficace des ressources du réseau. Dans cette thèse, nous considérons les éléments suivants qui compose de l'architecture pour la gestion dynamique des ressources : signalisation, algorithmes d'allocation de la bande passante et de contrôle de ressources, algorithmes de mesure des ressources dans le réseau, algorithmes de négociation entre plusieurs domaines. Nous proposons des critères de classification pour les protocoles QoS de signalisation en Internet et nous avons effectué une analyse des protocoles QoS les plus utilisés utilisant cette classification. Nous avons étudié l'impact entre les paramètres de SLA (bande passante, gigue, délai, perte des paquets) et les paramètres de différents mécanismes de QoS implémentés dans un réseau Linux Diffserv. En utilisant le réseau Linux Diffserv, nous montrons que la gestion statique de la QoS dans certains cas n'est pas efficace. Nous proposons un mécanisme d'allocation dynamique de la bande passante dans les routeurs à l'intérieur du réseau (Linux Diffserv). Avec notre algorithme nous proposons plusieurs règles de partage des ressources.

Techn. University, Diss., 2004--Darmstadt. / Dateien im PDF-Format.

Made available in DSpace on 2014-06-12T15:58:39Z (GMT). No. of bitstreams: 2 arquivo4665_1.pdf: 1244503 bytes, checksum: 9601c55f303bf34b27c25d55062533b9 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2003 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O Network Time Protocol (NTP) é um protocolo para sincronização de relógios de computadores em rede que tem quase duas décadas de existência e está em contínua evolução. Com algoritmos robustos para tratamento de fase e freqüência, com um bom disciplinador de relógio local e tendo implementações para as mais diversas plataformas e sistemas operacionais de computadores e equipamentos de rede, o NTP tornou-se um padrão de fato. Entretanto, esse protocolo ainda tem problemas que precisam ser solucionados para ser operacionalmente eficaz, pois seu desempenho depende de boas condições de rede para a troca das informações de sincronização, sofrendo bastante em caso de congestionamento. Devido a esses problemas, em vários países (incluindo o Brasil) esse protocolo é considerado inadequado para prover informação de horário de forma confiável, fazendo com que registros de hora em sistemas que o utilizam não tenham valor legal. Além disso, muitas aplicações necessitam de um sistema de controle de relógios confiável para funcionar corretamente (por exemplo, sistemas bancários e servidores de bancos de dados distribuídos). Isso obriga muitas empresas a utilizar sistemas legados tradicionais e mais caros para poder funcionar de forma correta e legal. Com o advento da Qualidade de Serviço em redes de computadores, esse problema pode ser abordado elegantemente e resolvido. Várias arquiteturas de Qualidade de Serviço foram propostas, mas a arquitetura de Serviços Diferenciados (DiffServ), devido a sua facilidade de implementação e ao fato de ter sido a mais estudada e implementada experimentalmente, mostrou-se a mais forte candidata à implantação mundial e no menor prazo. Essa arquitetura adequou-se bem ao problema de sincronização de relógios, embora a solução não seja trivial. Essa dissertação sugere um arcabouço para lidar com a sincronização de relógios através do NTP em domínios DiffServ com ou sem corretores de banda. Tendo-se Serviços Bem Definidos (Well Defined Services WDS) e baseado na idéia de que as aplicações devem conhecer o tipo de tratamento necessário ao seu tráfego, esse arcabouço consiste na adoção de políticas para o tratamento de tráfego NTP nos equipamentos de rede e na adoção de uma política para a marcação de pacotes por parte da aplicação. A proposta é validada com um estudo de caso feito com medição real do desempenho da aplicação sobre um ambiente de rede emulado

Over the last years, we have witnessed a rapid deployment of real-time applications on the Internet as well as many research works about Quality of Service (QoS) in particularly IPv4 (Internet Protocol version 4). The inevitable exhaustion of the remaining IPv4 address pool has become progressively evident. As the evolution of Internet Protocol (IP) continues, the deployment of IPv6 QoS is underway. Today, there is limited experience in the deployment of QoS for IPv6 traffic in MPLS backbone networks in conjunction with DiffServ (Differentiated Services) support. DiffServ itself does not have the ability to control the traffic which has been taken for end-to-end path while a number of links of the path are congested. In contrast, MPLS Traffic Engineering (TE) is accomplished to control the traffic and can set up end-to-end routing path before data has been forwarded. From the evolution of IPv4 QoS solutions, we know that the integration of DiffServ and MPLS TE satisfies the guaranteed QoS requirement for real-time applications. This thesis presents a QoS performance study of real-time applications such as voice and video conferencing over DiffServ with or without MPLS TE in IPv4/IPv6 networks using Optimized Network Engineering Tool (OPNET). This thesis also studies the interaction of Expedited Forwarding (EF), Assured Forwarding (AF) traffic aggregation, link congestion, as well as the effect of various performance metrics such as packet end-to-end delay, packet delay variation, queuing delay, throughput and packet loss. The effectiveness of DiffServ and MPLS TE integration in IPv4/IPv6 network is illustrated and analyzed. The thesis shows that IPv6 experiences more delay and loss performance than their IPv4 counterparts. / Author (1): Md. Tariq Aziz Address: C/O: Murshed Gularm, Nykarleby Gatan 134, 16474 Kista, Stockholm, Sweden Author (2): Mohammad Saiful Islam Address: Minnervavägen 22B, 371 43 Karlskrona, Sweden

Video streaming through IP networks has risen rapidly over the recent years, and will continue to do so over the coming years. In addition to this, new technologies such as Virtual Reality and robotics will lead to many new applications that will put high pressure on the networks. To combat these challenges, networks need to be application sensitive, and be able to provide Quality of Service (QoS) based on requirement. Network paradigms like Software Defined Networking (SDN) enables the network to be directly programmable, and could thus solve the challenge. In this thesis, the objective is to research if SDN can provide High QoS. Methods are developed to achieve High QoS with SDN. A combination of Differentiated Services Code Point (DSCP) values and DSCP remarking with Meters are used enable High QoS and together they can give bandwidth guarantee. As a result of the thesis, a solid theoretical method is provided for achieving QoS, tests are performed and show that QoS can be implemented in SDN, but it is unable to implement High QoS due to the lack of implementation for Meters with DSCP remarking.

The aim of this diploma thesis was to create an own DiffServ domain model, which is supplemented with stations generating various type of network traffic and implementation of token-bucket mechanism in router´s process model in Opnet Modeler simulation environment. DiffServ domain is made up from two edge and two core routers and includes servers and client stations generating VoIP, FTP, HTTP and database access traffic. It is described a process of distribution of traffic into different classes on edge routers of DiffServ domain along with assurance of separate handling with usage of an Assured Forwarding PHB mechanism. In point of differentiated packet processing is process model completed with generating of various statistics. The process of their creating is used to check a dropper activity, which is implemented together with token-bucket mechanism on ARP layer of the router. The functionality of the model is verified by simulation.

Information about the differentiated services (DiffServ), the link failure problems and the alternate routes in a DiffServ domain finding are the aims of the master´s thesis to be given. Properties of the differentiated services, the usage of marking techniques in the DiffServ domain are described in the thesis. The concept of the DiffServ domain, its structure and mostly used today routing protocol OSPF are described too. The function model of the DiffServ domain has been created in the simulation environment called Opnet Modeleler on the basis of the obtained knowledge. The link failure and the following link recovery have been simulated in the model and the effects have been analyzed after that.