Esta dissertação apresenta a construção de um protótipo de servidor Web distribuído, baseado no modelo de servidor Web com diferenciação de serviços (SWDS) e a implementação e avaliação de algoritmos de seleção, utilizando o conceito de controle de congestionamento para requisições HTTP. Com isso, além de implementar uma plataforma de testes, este trabalho também avalia o comportamento de dois algoritmos de controle de congestionamento. Os dois algoritmos estudados são chamados de Drop Tail e RED (Random Early Detection), no qual são bastante difundidos na literatura científica e aplicados em redes de computadores. Os resultados obtidos demostram que, apesar das particularidades de cada algoritmo, existe uma grande relação entre tempo de resposta e a quantidade de requisições aceitas / This MSc dissertation presents the implementation of a prototype for a distributed web server based on the SWDS, a model for a web server with service differentiation, and the implementation and evaluation of selection algorithms adopting the concept of congestion control for HTTP requests. Thus, besides implementing a test platform this work also evaluates the behavior of two congestion control algorithms. The two algorithms studied are the Drop Tail and the RED (Random Early Detection), which are frequently discussed in the scientific literature and widely applied in computer networks. The results obtained show that, although the particularities of each algorithm, there is a strong relation between the response times and the amount of requests accepted in the server

A Qualidade de Serviço oferece aos Sistemas de Tempo Real garantias das restrições temporais de aplicações tais como comércio eletrônico, vídeo conferência, telemedicina, entre outras que necessitam de confiabilidade e desempenho para efetivação dos seus serviços. Contribuições nessas áreas têm focado a provisão de qualidade de serviço em termos absolutos, descritos num limite máximo de média de tempo de atendimento, destinado às aplicações Web. Todavia, o controle de qualidade baseado em tempo médio de resposta durante todo o período de serviço prestado, não restringe os tempos de resposta quanto a sua dispersão. Um valor médio razoável pode ser resultado da combinação de tempos de resposta elevados e pequenos. O objetivo deste trabalho é reduzir a dispersão dos atrasos em sistemas escalonados com a política EBS (Exigency Based Scheduling ), definidos pela diferença positiva entre o tempo de resposta da requisição do usuário e o valor médio contratado, em um modelo de contrato determinado por janelas deslizantes de operações. A EBS é uma política de escalonamento que define um modelo de contratos de tempo médio de resposta avaliados durante todo o tempo de serviço. Avalia-se a dispersão dos atrasos, pois, é preferível que haja a possibilidade dos tempos de resposta ficarem substancialmente abaixo da média contratada. Os resultados são obtidos utilizando métodos de simulação dos algoritmos desenvolvidos durante o projeto / Quality of Service offers Real Time Systems guarantees the time constraints of applications such as electronic commerce, video conferencing, telemedicine, among others that need reliable and effective performance for their services. Contributions in these areas have made the provision of service quality in absolute terms, out of a maximum average time of service for Web applications. However, quality control based on average response time for the entire period of service, does not restrict the response time to their dispersion. A reasonable average value may be the result of the combination of response times of high and low. Objective of this research is reduce the delays dispersion in systems scheduled with policy EBS (Exigency Based Scheduling), defined by the positive difference between the response time of user request and the average contract value, in a model contract by a sliding operation windows. The EBS is a scheduling policy that defines a contracts model for an average response time evaluated throughout the time of service. We analyze the delays dispersion because it is preferable that there is the possibility of response time becomes substantially below the average contracted. Results are obtained using simulation methods of the algorithms developed during the project

O aumento no número de dispositivos móveis, como smartphones, tablets e laptops, e o avanço em seu potencial computacional permitiu considerá-los como recursos computacionais. O uso de recursos computacionais com maior proximidade vem crescendo ano após ano, sendo chamado de Fog computing, em que os elementos na borda da Internet são explorados, uma vez que os serviços computacionais convencionais podem estar indisponíveis ou sobrecarregados. Dessa forma, este projeto de Mestrado tem como foco possibilitar o uso de dispositivos móveis no provimento de serviços computacionais entre si de forma colaborativa através da heurística Maximum Regret adaptada, que busca alocar tarefas computacionais em dispositivos locais de forma a minimizar o consumo de energia e evitar dispositivos não confiáveis. Também há uma meta-heurística em um nível global, que interconecta os diferentes aglomerados de dispositivos móveis na borda da Internet, e possui informações globais de Quality of Service (QoS). Foram realizados experimentos que mostraram que evitar dispositivos móveis como recursos com um baixo grau de confiabilidade possibilitou diminuir o impacto no consumo de energia, além de ser possível diminuir os tempos de resposta e de comunicação ao ajustar a política de seleção de aglomerados externos. / The increasing number of mobile devices, such as smartphones, tablets and laptops, as well as advances in their computing power have enabled us to consider them as resources, exploring the proximity. The use of near computing resources is growing year by year, being called as Fog computing, where the elements on the edge of the Internet are exploited, once the computer services providers could be unavailable or overloaded. Thus, this Masters project focuses on using mobile devices to provide computing services among them through a heuristic called Adapted Maximum Regret, which tries to minimize energy consumption and avoid untrustable devices. There is also top-level metaheuristic which interconnects different clusters of devices on the edge of the Internet with global information to guarantee Quality of Services (QoS). We conducted a set of experiments that showed us to avoid devices with a high degree of failures to save more energy when allocating tasks among them, as well as decreasing the applications response time and communication through adjusts in the selection algorithm of external agglomerates.

Esta tese de doutorado apresenta a investigação de alternativas para melhorar o desempenho de ambientes Web, avaliando o impacto da utilização de mecanismos de diferenciação de serviços em todos os pontos do sistema. Foram criados e modelados no OPNET Modeler cenários com diferentes configurações voltadas tanto para a diferenciação de serviços, quanto para o congestionamento da rede. Foi implementado um servidor cache com suporte à diferenciação de serviços (cache CDF), que constitui uma contribuição dentro deste trabalho, complementando o cenário de diferenciação de serviços de forma positiva, assegurando que os ganhos obtidos em outras etapas do sistema não sejam perdidos no momento da utilização do cache. Os principais resultados obtidos mostram que a diferenciação de serviços introduzida de forma isolada em partes do sistema, pode não gerar os ganhos de desempenho desejados. Todos os equipamentos considerados nos cenários propostos possuem características reais e os modelos utilizados no OPNET foram avaliados e validados pelos seus fabricantes. Assim, os modelos que implementam os cenários considerados constituem também uma contribuição importante deste trabalho, uma vez que o estudo apresentado não se restringe a uma modelagem teórica, ao contrário, aborda aspectos bem próximos da realidade, constituindo um possível suporte de gerenciamento de sistemas Web / This PhD thesis presents the investigation of alternatives to improve the performance of Web environments by evaluating the impact of using differentiated service mechanisms in all points of the system. Several scenarios were created and modeled in the OPNET Modeler, with different configurations of both differentiated services and network overloading. A special cache server supporting differentiated services (CDF cache) was proposed and included in the model, comprising one of the major contributions of this work once it positively complements the differentiated service scenario, making that the gains obtained with other stages of the system do not be spoiled when using the cache. The main results obtained show that the adoption of differentiated services in isolated parts of the system cannot generate the expected performance gains. The features of all the equipments considered in the several scenarios defined in this work are very close to the reality and the models used in the OPNET were evaluated and validated by the companies that produce those equipments. Thus, the models that implement the scenarios considered in this work also comprises an important contribution of this thesis, once the study presented is not just a theoretical modeling exercise but, conversely, it approaches aspects very close to the reality, comprising a possible Web system management support

Submitted by PPG Ci?ncia da Computa??o (ppgcc@pucrs.br) on 2018-04-03T14:37:48Z No. of bitstreams: 1 MARCELO_RUARO_TES.pdf: 4683751 bytes, checksum: 6eb242e44efbbffa6fa556ea81cdeace (MD5) / Approved for entry into archive by Tatiana Lopes (tatiana.lopes@pucrs.br) on 2018-04-13T17:30:40Z (GMT) No. of bitstreams: 1 MARCELO_RUARO_TES.pdf: 4683751 bytes, checksum: 6eb242e44efbbffa6fa556ea81cdeace (MD5) / Made available in DSpace on 2018-04-13T17:37:13Z (GMT). No. of bitstreams: 1 MARCELO_RUARO_TES.pdf: 4683751 bytes, checksum: 6eb242e44efbbffa6fa556ea81cdeace (MD5) Previous issue date: 2018-03-16 / Sistemas multi-n?cleos intra-chip s?o o estado-da-arte em termos de poder computacional, alcan?ando de d?zias a milhares de elementos de processamentos (PE) em um ?nico circuito integrado. Sistemas multi-n?cleos de prop?sito geral assumem uma admiss?o din?mica de aplica??es, onde o conjunto de aplica??es n?o ? conhecido em tempo de projeto e as aplica??es podem iniciar sua execu??o a qualquer momento. Algumas aplica??es podem ter requisitos de tempo real, requisitando n?veis de qualidade de servi?o (QoS) do sistema. Devido ao alto grau de imprevisibilidade do uso dos recursos e o grande n?mero de componentes para se gerenciar, propriedades autoadaptativas tornam-se fundamentais para dar suporte a QoS em tempo de execu??o. A literatura fornece diversas propostas de QoS autoadaptativo, focado em recursos de comunica??o (ex., redes intra-chip), ou computa??o (ex., CPU). Contudo, para fornecer um suporte de QoS completo, ? fundamental uma autoconsci?ncia abrangente dos recursos do sistema, e assumir t?cnicas adaptativas que permitem agir em ambos os n?veis de comunica??o e computa??o para atender os requisitos das aplica??es. Para suprir essas demandas, essa Tese prop?e uma infraestrutura e t?cnicas de gerenciamento de QoS autoadaptativo, cobrindo ambos os n?veis de computa??o e comunica??o. No n?vel de computa??o, a infraestrutura para QoS consiste em um escalonador din?mico de tarefas de tempo real e um protocolo de migra??o de tarefas de baixo custo. Estas t?cnicas fornecem QoS de computa??o, devido ao gerenciamento da utiliza??o e aloca??o da CPU. A novidade do escalonador de tarefas ? o suporte a requisitos de tempo real din?micos, o que gera mais flexibilidade para as tarefas em explorar a CPU de acordo com uma carga de trabalho vari?vel. A novidade do protocolo de migra??o de tarefas ? o baixo custo no tempo de execu??o comparado a trabalhos do estado-da-arte. No n?vel de comunica??o, a t?cnica proposta ? um chaveamento por circuito (CS) baseado em redes definidas por software (SDN). O paradigma SDN para NoCs ? uma inova??o desta Tese, e ? alcan?ado atrav?s de uma arquitetura gen?rica de software e hardware. Para QoS de comunica??o, SDN ? usado para definir caminhos CS em tempo de execu??o. Essas infraestruturas de QoS s?o gerenciadas de uma forma integrada por um gerenciamento de QoS autoadaptativo, o qual segue o paradigma ODA (Observar, Decidir, Agir), implementando um la?o fechado de adapta??es em tempo de execu??o. O gerenciamento de QoS ? autoconsciente dos recursos do sistema e das aplica??es em execu??o, e pode decidir por adapta??es no n?vel de computa??o ou comunica??o, baseado em notifica??es das tarefas, monitoramento do ambiente, e monitoramento de atendimento de QoS. A autoadapta??o decide reativamente assim como proativamente. Uma t?cnica de aprendizagem do perfil das aplica??es ? proposta para tra?ar o comportamento das tarefas de tempo real, possibilitando a??es proativas. Resultados gerais mostram que o gerenciamento de QoS autoadaptativo proposto pode restaurar os n?veis de QoS para as aplica??es com um baixo custo no tempo de execu??o das aplica??es. Uma avalia??o abrangente, assumindo diversos benchmarks mostra que, mesmo sob diversas interfer?ncias de QoS nos n?veis de computa??o e comunica??o, o tempo de execu??o das aplica??es ? restaurado pr?ximo ao cen?rio ?timo, como 99,5% das viola??es de deadlines mitigadas. / Many-core systems-on-chip are the state-of-the-art in processing power, reaching from a dozen to thousands of processing elements (PE) in a single integrated circuit. General purpose many-cores assume a dynamic application admission, where the application set is unknown at design-time and applications may start their execution at any moment, inducing interference between them. Some applications may have real-time constraints to fulfill, requiring levels of quality of service (QoS) from the system. Due to the high degree of resource?s utilization unpredictability and the number of components to manage, self-adaptive properties become fundamental to support QoS at run-time. The literature provides several self-adaptive QoS proposals, targeting either communication (e.g., Network-on-Chip) or computation resources (e.g., CPU). However, to offer a complete QoS support, it is fundamental to provide a comprehensive self-awareness of the system?s resources, assuming adaptive techniques enabling to act simultaneously at the communication and computation levels to meet the applications' constraints. To cope with these requirements, this Thesis proposes a self-adaptive QoS infrastructure and management techniques, covering both the computation and communication levels. At the computation level, the QoS-driven infrastructure comprises a dynamic real-time task scheduler and a low overhead task migration protocol. These techniques ensure computation QoS by managing the CPU utilization and allocation. The novelty of the task scheduler is the support for dynamic real time constraints, which leverage more flexibility to tasks to explore the CPU according to a variable workload. The novelty of the task migration protocol is its low execution time overhead compared to the state-of-the-art. At the communication level, the proposed technique is a Circuit-Switching (CS) approach based on the Software Defined Networking (SDN) paradigm. The SDN paradigm for NoCs is an innovation of this Thesis and is achieved through a generic software and hardware architecture. For communication QoS, SDN is used to define CS paths at run-time. A self-adaptive QoS management following the ODA (Observe Decide Act) paradigm controls these QoS-driven infrastructures in an integrated way, implementing a closed loop for run time adaptations. The QoS management is self-aware of the system and running applications and can decide to take adaptations at computation or communication levels based on the task feedbacks, environment monitoring, and QoS fulfillment monitoring. The self-adaptation decides reactively as well as proactively. An online application profile learning technique is proposed to trace the behavior of the RT tasks and enabling the proactive actions. Results show that the proposed self-adaptive QoS management can restore the QoS level for the applications with a low overhead over the applications execution time. A broad evaluation, using known benchmarks, shows that even under severe QoS disturbances at computation and communication levels, the execution time of the application is restored near to the optimal scenario, mitigating 99.5% of deadline misses.

Esta dissertação de mestrado aborda um estudo sobre a caracterização de cargas de trabalho para Web services por meio da análise de Documentos WSDL (Web Service Description Language). Esses documentos, que representam a interface para os serviços, foram obtidos na Web e seus conteúdos analisados estatisticamente. A metodologia utilizada para alcançar a caracterização desejada, constitui-se da coleta e análise dos dados de 1346 arquivos WSDL válidos. O resultado do estudo proposto nesta dissertação contribui para a avaliação de desempenho no que diz respeito a caracterizar os diversos aspectos das cargas de trabalho de Web services. Dentre os dados obtidos ressaltam-se as porcentagens da ocorrência de algumas características tais como o número de operações, tipos de Binding, quantidade de parâmetros de entrada e saída e tipos de dados mais utilizados. Para auxiliar na busca e avaliação das características das WSDLs uma nova ferramenta denominada WSDLAnalyzer é proposta e implementada / This MSc dissertation approaches a study on workload characterization for web services by means of the analysis of WSDL (Web Service Description Language) documents. These documents, that represent the interface to the services, were all obtained in the web and their contents were statistically analyzed. The methodology used to reach the desired characterization comprises the collection and analysis of 1346 WSDL valid files. The result of the study proposed in this dissertation contributes to the performance evaluation of web services in terms of characterizing the several aspects of the workloads. Among the data obtained, the percentage of occurrences of some features, such as the number of operations, the types of binding, the amount of input and output parameters and the more used data types, are highlighted. To support the search and evaluation of the WSDL features a new tool, named WSDLAnalyzer, is proposed and implemented

A grande evolução nos sistemas de comunicação, relacionados à transmissão de conteúdo multimídia nas redes de comunicação de dados e de transmissão de TV, demanda soluções de gerência da qualidade de serviço fim-a-fim. Diante deste cenário é proposta nesse trabalho a definição, implementação e validação de uma solução para monitoramento de transmissão de mídia que permite ser aplicada em arquiteturas de redes heterogêneas. Essas redes possuem a característica de integrar diversas tecnologias de redes de comunicação de dados e de telecomunicações, envolvendo redes modernas e antigas, padronizadas e proprietárias. Este trabalho apresenta um sistema de monitoramento para transmissão de TV Digital em ambientes heterogêneos convergentes, isto é, redes que utilizam tecnologias e meios de transmissão distintos para transmissão de vários tipos de fluxos, e com foco na aplicação TV Digital. Para isso, são discutidos aspectos fundamentais para o monitoramento de fluxo de TV Digital neste tipo de rede e é feita uma análise da comunicação entre os subsistemas do ambiente. O sistema de monitoramento possui uma arquitetura definida que se baseia no protocolo RTP Real-time Transport Protocol, para suportar a transmissão de mídias, como áudio e vídeo, em redes heterogêneas. A fim de validar esta arquitetura foi implementado um protótipo, como prova de conceito, que obtém dados através da base de informações de gerência (MIB) RTP. Os resultados gerados pelo protótipo consolidam informações relativas à variação de atraso em uma interface web de gerenciamento através de monitores distribuídos pela rede. / The great evolution in the telecommunication systems related to the transmission of multimidia contents in the data communication networks and TV transmission, demands management solutions in which regards quality of end-to-end services. Taking this scenario into consideration, it is proposed in this work the definition and implementation of a solution to monitor media transmition that can be applied to heterogeneous network architectures. The feature of such networks is to integrate different data communication network technologies and telecommunication, involving legacy and modern networks, standarized and proprietary. This work presents a monitoring system for digital TV transmission in heterogeneous converging environments, that is, networks that use different technologies and transmission means for the Digital TV application. Therefore, key aspects for monitoring the digital TV flow in this kind of network are pointed out, as well as an analysis of the communication among the environment subsystems. The monitoring system has a defined architecture, based on the RTP Real-time Transport Protocol, to support the media transmission, such as audio and video, in heterogeneous networks. In order to validate this architecture, a prototype was implemented, as concept evidence, that obtains data from the management information base (MIB) RTP. The results generated by the prototype consolidate information related to the delay variation (jitter) in a management web interface through monitors distributed around the network.

Nowadays, the main concern for IT-related service sector in organizations is to improve IT-Service Management (ITSM) by reducing cost of service in conjunction with growing in their efficiency and effectiveness. Particularly in IT firms, these concerns are much more imperative where the ITSM framework is the main section of providing service and feedback to managers. However, a challenge is that this significant service segment is ineffective by customers (I.e. IT managers) since IT maintenance and IT operations are unnoticed by IT related research. Academia together with industry demand a better knowledge and information of IT service management to work closely in order to improve efficiency and effectiveness of the system function.In this research study, it is attempted to review the situation of ITSM framework (I.e., Metrics) about evaluation method based on Key Performance Indicators (KPIs) and best practices for IT-Service Management process in DB Schenker and to compare them with other corporations. At first, the researcher has done an extensive literature review on the earlier works of Metrics for IT Service Management. After extracting needed information, it is governed in an innovative way to make it more effective and efficient, then run a case study for better understanding of the real situation and to compare literature and real world. The researchers mainly choose Swedish companies as successful companies in IT management to review ITSM framework (i.e. KPIs). / Program: Masterutbildning i Informatik

Les communautés virtuelles ont de plus en plus d’influence dans nos activités quotidiennes. Qu’elles soient sociales, d’affaires, professionnelles, d’apprentissage, ces communautés sont en concurrence pour la conquête de l'Internet, en ciblant une audience de plus en plus large et en proposant une offre de services de plus en plus variée. Par voie de conséquence, le succès ou l'échec de ces communautés dépend largement des services proposés dont la diversité, la qualité et l'adaptation sont les facteurs clés de satisfaction des clients. C’est pourquoi la démarche SOA (Service Oriented Architecture /Architecture Orientée Service) favorise la vision d'environnements ouverts où services, fournisseurs et clients sont indépendants les uns des autres, grâce au découplage et à l'allocation dynamique des services. Malheureusement, les environnements de communautés virtuelles ne prennent pas vraiment en compte les principes SOA et sont considérés fermés d’un point de vue des services offerts car ceux-ci sont limités aux fonctionnalités de la plateforme qui les hébergent. Cette dépendance des services vis-à-vis de la plateforme est considérée comme une limitation qui influence d'une manière négative le succès et la durabilité des communautés virtuelles. Du point de vue des membres d’une communauté, cette limitation entraine le départ de certains d’entre eux, et/ou impose à ses membres de joindre d'autres communautés afin de bénéficier des services offerts par ces dernières qui ne sont pas disponibles dans leur communauté d'origine. Du point de vue de l’environnement, l'introduction de nouveaux services nécessite de modifier la plateforme existante, et peut demander dans certains cas une migration vers une autre plateforme, ce qui peut perturber la communauté en question lorsque celle-ci est opérationnelle avec des membres en ligne. Dans ce contexte, ce travail a pour but de palier les limites de la gestion de services dans les communautés virtuelles afin de satisfaire les besoins de leurs membres, d'assurer une meilleure gestion des services d'un point de vue individuel et d'un point de vue de la communauté, et de garantir une évolution dynamique des services au sein de la communauté. L’objectif principal est donc de " Fournir le bon service, au bon utilisateur, au bon moment et avec la bonne qualité". L’hypothèse fondatrice de ce travail est que les communautés virtuelles peuvent être construites en commençant par un ensemble minimal de services de base, cet ensemble pouvant ensuite être étendu par l'ajout de nouveaux services selon les besoins des membres de la communauté. En adoptant cette approche, nous proposons un cadre de gestion de services qui aborde les difficultés rencontrées par les communautés virtuelles et leurs membres. En conséquence, le focus porte sur la satisfaction de ces membres plutôt que sur le service lui-même ou le fournisseur du service. Ainsi, nous définissons une nouvelle structuration des services au sein d’une communauté qui s’appuie sur une classification en différentes catégories fonctionnelles. Puis, nous étendons l'architecture SOA avec les concepts nécessaires pour modéliser ces catégories et leur associer un ensemble de propriétés non fonctionnelles de Qualité de Service (QdS ou QoS en anglais) utilisées par un système de médiation pour proposer les services adaptés aux besoins des usagers. Une description des unités fonctionnelles de ce système, ainsi que la façon dont elles opèrent, coopèrent et collaborent afin d'accomplir l’objectif défini ci-dessus constitue le cœur de notre contribution. / Virtual Communities are dominating our daily activities from different insights. Social, Business, Professional, Educational and many virtual communities are competing among each other to conquer the internet by targeting more audience through the services they provide. Consequently, the success or failure of virtual communities depends to a great extent on its services. In a world driven by services, diversity, quality and adaptation are key factors to achieve customer satisfaction. Accordingly the Service Oriented Architecture (SOA) approach promotes the vision of open environments where services, providers and consumers are considered independently from one another thanks to decoupling and dynamic allocation of services. But virtual communities environment did not really care about SOA and are considered closed with respect to the services they provide since they are bounded to the capabilities of the platform that host them. This implies the delivery of services inside the virtual communities is dependent on the platform used which is considered a limitation that have negative influence on the success and sustainability of virtual communities. From a member perspective this limitation causes community members to leave the community, and/or imposes them to join other virtual communities to benefit from the services they host and that are not available in their home virtual communities. From an environment perspective, introducing new services into these communities require modifications on the existing platforms or might require a complete shift to another platform in some cases which might affect the target community in case it is operational with active users. In this context, our research work aims to overcome the limitation in managing services of virtual community to satisfy community members’ needs, to provide better service management from a member perspective as well as from a community perspective, and to guarantee dynamic evolution of services inside the community. Our main objective is “To provide the right service to the right user in the right time with the required quality of service”. Our assumption is that virtual communities can be built starting from a minimal set of basic services and then add more services based on the needs of the community members. This drives us to adopt this approach and propose a service management framework that address the challenges faced by virtual communities and their members. Accordingly, we approach the problem from a members’ perspective and choose to work on members’ satisfaction more than we care about the service itself or the provider of the service. Thus, we define a new structure of services within a community that is based on a classification into different functional categories. Then, we extend SOA with the concepts necessary to model these categories and associate a set of non-functional properties of Quality of Service (QoS ) used by a mediation system to offer services best suited to the needs of members. Finally, we provide a description of the functional units of the system and how they operate, cooperate and collaborate to achieve the aforementioned objective. This is the core of our contribution.

Esta dissertação apresenta a construção de um protótipo de servidor Web distribuído, baseado no modelo de servidor Web com diferenciação de serviços (SWDS) e a implementação e avaliação de algoritmos de seleção, utilizando o conceito de controle de congestionamento para requisições HTTP. Com isso, além de implementar uma plataforma de testes, este trabalho também avalia o comportamento de dois algoritmos de controle de congestionamento. Os dois algoritmos estudados são chamados de Drop Tail e RED (Random Early Detection), no qual são bastante difundidos na literatura científica e aplicados em redes de computadores. Os resultados obtidos demostram que, apesar das particularidades de cada algoritmo, existe uma grande relação entre tempo de resposta e a quantidade de requisições aceitas / This MSc dissertation presents the implementation of a prototype for a distributed web server based on the SWDS, a model for a web server with service differentiation, and the implementation and evaluation of selection algorithms adopting the concept of congestion control for HTTP requests. Thus, besides implementing a test platform this work also evaluates the behavior of two congestion control algorithms. The two algorithms studied are the Drop Tail and the RED (Random Early Detection), which are frequently discussed in the scientific literature and widely applied in computer networks. The results obtained show that, although the particularities of each algorithm, there is a strong relation between the response times and the amount of requests accepted in the server