Proposta e avaliação: algoritmos para gerência de banda e seleção de caminhos em redes MPLS (DS-TE)

• Ano de defesa: 2008
• Autor(a) principal: Pinto Neto, Walter da Costa
• Orientador(a): Martins, Joberto Sérgio Barbosa
• Banca de defesa: Monteiro, José Augusto Suruagy , Souza, José Neuman
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Salvador
• Programa de Pós-Graduação: Programa de Pós-Graduação em Sistemas e Computação
• Departamento: Sistemas e Computação
• País: BR
• Palavras-chave em Português: DiffServ Aware-MPLS Traffic Engineering (DS-TE); Russian Doll Model (RDM); Modelo de Restrição de Banda; Engenharia de Tráfego; Qualidade de Serviço; Avaliação de Desempenho de Redes de Computadores; Simulação
• Palavras-chave em Inglês: DiffServ Aware MPLS Traffic Engineering (DS-TE); Russian Doll Model (RDM); Bandwidth Constraint Model; Traffic Engineering; Quality of Service; Performance Evaluation of Computer Networks; Simulation
• Área do conhecimento CNPq: CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::TEORIA DA COMPUTACAO::ANALISE DE ALGORITMOS E COMPLEXIDADE DE COMPUTACAO
• Link de acesso: http://teste.tede.unifacs.br:8080/tede/handle/tede/258

This work aims at proposing and evaluating a bandwidth sharing management algorithm among class types and a path selection algorithm for DiffServ Aware-MPLS Traffic Engineering (DS-TE) networks configured with RDM (Russian Doll Model) bandwidth constraint model. Initially, the technologies that offer Quality of Service (QoS) and Traffic Engineering (TE) in multiservice IP networks, together with the DS-TE architecture are introduced. The ADAPT-RDM and MinpreptDSTE algorithms are presented and the results of a performance evaluation done by simulation are discussed for different topologies networks scenarios. The simulation was done by a specific simulator developed in the scope of this work, with the specific proposal of evaluate the performance of the proposed algorithms. The results show that the ADAPT-RDM algorithm preserves the RDM model configuration for networks links independently of LSPs path selection criteria and network traffic level. The LSP path selection resulted by use of the MinpreptDSTE algorithm reduced the total number of LSPs preemptions, when compared with the minimum hops criteria used by the Constrained Shortest Path First (CSPF) algorithm