Made available in DSpace on 2014-07-29T15:08:15Z (GMT). No. of bitstreams: 1 LeonardoAntonio alves.pdf: 913407 bytes, checksum: 5f344f91c6101d278dcd0a211a753de0 (MD5) Previous issue date: 2010-08-19 / In this work, we present an analytical model to predict the performance of the client-server system regarding an E-Learning platform. This approach is based on the queuing theory and can adequately describe the performance of an E-Learning platform and its parameters, such as the server response time. It is used a Poisson model to describe the traffic processes in the EaD system, comparing the output of the model to an E-Learning platform working at the MSD Education company. In this study, it was developed a precise tool for designing E-Learning platforms based on client-server model according to the increasing access demand. / Neste trabalho, propõe-se um modelo de análise para prever o desempenho de um sistema cliente-servidor referente a uma plataforma de Educação a Distância. Esta abordagem é baseada na teoria de filas e pode descrever adequadamente o desempenho de uma plataforma de Educação a Distância e seus parâmetros, como por exemplo, o tempo médio de resposta do servidor. Utiliza-se o modelo de Poisson para descrever o processo de requisições no sistema EaD, comparando os resultados fornecidos pelo modelo com os de uma plataforma de Educação a Distância mantida pela empresa MSD Educação. Neste estudo, desenvolveu-se uma ferramenta precisa para a concepção e dimensionamento de plataformas de Educação a Distância baseada no modelo cliente-servidor de acordo com a demanda crescente de acesso.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / A presente dissertaÃÃo de mestrado trata do desenvolvimento teÃrico-experimental de um novo modelo da Turbina Tesla, tambÃm conhecida como turbina de discos, em que o foco da pesquisa foi estudar esse dispositivo de propulsÃo visando sua aplicaÃÃo em sistemas de geraÃÃo de energia distribuÃda a partir de fontes renovÃveis de biomassa. Um protÃtipo da turbina foi desenvolvido, em escala de laboratÃrio, para determinar os parÃmetros de projeto atravÃs de estudos teÃrico e experimental, a fim de interpretar os fenÃmenos presentes nos processos internos desse tipo de mÃquina de fluxo. A pesquisa foi direcionada à anÃlise dos efeitos do espaÃamento entre discos sobre o torque e a potÃncia, como uma funÃÃo da camada limite fluidodinÃmica do escoamento, correlacionando os regimes de velocidades de escoamento Ãs propriedades do fluido de trabalho. A metodologia para cÃlculo das eficiÃncias isentrÃpicas e exergÃticas à estabelecida a partir de dados experimentais e da anÃlise teÃrica por simulaÃÃo (apoiada pelo software ANSYS-CFX), a qual determina o perfil de distribuiÃÃo de pressÃo dentro da cÃmara de expansÃo, caracterizando as linhas de escoamento e os gradientes de pressÃo. Complementarmente, uma anÃlise modal dos sinais de vibraÃÃo da turbina foi realizada, estudando o comportamento dinÃmico da estrutura com o propÃsito de detectar as freqÃÃncias naturais e investigando a influÃncia do espaÃamento dos discos sobre o nÃvel de vibraÃÃodesse dispositivo. Este estudo aponta a robustez e a versatilidade operacional desse tipo de turbina, pelo fato dela admitir ampla faixa de rotaÃÃo e por sua fÃcil adaptaÃÃo a geradores comerciais, sugerindo sua aplicaÃÃo em sistemas de geraÃÃo e co-geraÃÃo tanto para o meio rural como para a indÃstria. / The current masterâs dissertation treats the experimental and theoretical development of a new model of Tesla turbine, also known as disk turbine, in which the research focus was to study this propulsion device viewing its application in distributed energy generation systems from biomass renewable resources. A turbineâs prototype was developed, in laboratory scale, to determine the design parameters through the theoretical and experimental studies in order to interpret the phenomena present in the internal processes of this kind of flux machine. The research was directed toward the effect analysis of the gap between disks on the torque and the power, as a function of the flowâs fluidynamic boundary layer, correlating the flow velocity regimes with the working fluid properties. A methodology for isentropic and exergetic efficiency calculation is taken from experimental data and theoretical analyses through simulation (supported by the software ANSYS-CFX), which determines the pressure distribution profile inside the expansion chamber, characterizing flow lines and pressure gradients. Complementarily, a modal analysis of turbineâs vibration signals is performed, studying the dynamic behavior of the structure with the purposes of detecting the natural frequencies and investigating the influence of the diskâs gap on the vibration level of this device. This study points out the robustness and the operational versatility of this kind of turbine by the fact of it allows a wide rotation range and by its easy adaptation to commercial generators, suggesting its application into generation and co-generation systems for the rural area as well as for the industry.

O objetivo deste trabalho é modelar a arquitetura de uma usina solar fotovoltaica arrefecida intitulada UFVa, utilizando um protótipo de verificação. A metodologia se baseia na medição, verificação e análise dos dados de temperatura e produção de energia elétrica dos strings de teste (arrefecido) e comparação (não arrefecido), estudo do comportamento da alimentação de água do sistema de arrefecimento e o impacto das condições climáticas na operação do protótipo de UFVa. Por meio das análises dos dados constatou-se que, para o período entre as 09h00min e as 17h30min, os módulos PV do string de teste sempre operam com temperaturas inferiores aos módulos PV do string de comparação. Durante o período de testes, no qual a temperatura dos módulos PV do string de comparação operou acima de 55,0°C, as temperaturas médias e máximas registradas nos módulos PV do string de teste foram inferiores a 37,0 °C, operando sempre abaixo da temperatura nominal de operação da célula (NOCT). A produção de energia elétrica no string de teste superou a do string de comparação em 3,0kWh/dia. Portanto, o sistema de arrefecimento reduz a temperatura de operação dos módulos PV, principalmente no período de máxima geração elétrica, que corresponde ao período das 11h00min às 15h00min, proporcionando ganhos médios de rendimento de 5,9% na produção de energia, 10,3% na potência e 5,4% no FC. / In this work we use a verification prototype to model the architecture of a solar photovoltaic power plant equipped with a cooling system. The power plant we model is called UFVa. The methodology is based on the measurement, verification, and data analysis of temperature, electricity generation, test strings (cooled) and comparison strings (not cooled), along with a study of the water feeding behavior of the cooling system, and the impact of climatic conditions in the UFVa prototype operation. By analyzing the data we observed that, for the period between 09:00am and 5:30pm, the PV modules of the test string operate at temperatures below those of the PV modules of the comparison string. During the tests, in which the temperature of the PV modules of the comparison string operated above 55.0°C, the average and the maximum temperatures recorded in the PV modules of the testing string lied below 37.0°C, operating below the NOCT. Regarding the generation of electricity, the test string generated 3.0 kWh/day more than the comparison string. Hence, the cooling system decreases the operating temperature of the PV modules, particularly during the maximum power generation period which is from 11am to 3pm. This leads to efficiency average gains of up to 5.9% in the generation of electricity, 10.3% in the power, and 5.3% in the PR and PF.

Orientador: Walter da Cunha Borelli / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-04T00:34:51Z (GMT). No. of bitstreams: 1 Castro_MarcelCavalcantide_M.pdf: 1488533 bytes, checksum: 5a2e57ba0fe96d5603c60ef84835b558 (MD5) Previous issue date: 2004 / Resumo: Este trabalho apresenta propostas de implementação de qualidade de serviço na arquitetura VPN MPLS e analise de desempenho destas propostas. São desenvolvidos sistemas com base na arquitetura VPN MPLS e sugerido uma proposta de expansão da arquitetura VPN MPLS para mapeamento dinamico das prioridades dos clientes VPN na rede do provedor de serviço atraves da inserção do valor de prioridade de rota na tabela vrf e modificações realizadas no protocolo MP-BGP. As propostas foram especificadas utilizando a linguagem de especificação formal SDL orientada a objetos a partir da ferramenta SDL TAU Suite. Esta ferramenta permite simular os sistemas especificados, a partir de diagramas MSC, e validar estes sistemas para detecção e correção dos erros de logica e de especificação. A analise de desempenho das propostas foi realizada com o uso do simulador Opnet Modeler / Abstract: This work describe proposals for the implementation of quality of service (QoS) in VPN MPLS architecture, and their performance analysis based on simulations are presented. New systems are developed based on the VPN MPLS architecture, and it is being proposed an extension for the VPN MPLS architecture to construct a dynamic mapping of VPN clients priorities into service provider network through the insertion of priority field at vrf table and MP-BGP protocol modification. All these systems were specified using the SDL object-oriented formal language with the SDL TAU Suite tool (Telelogic, Sweden). This tool allows the simulation with MSC diagrams of the specified systems, and the validation of these systems by detecting and correcting logical and specification errors. The performance analysis of these proposals were realized through the Opnet Modeler simulator / Mestrado / Telecomunicações e Telemática / Mestre em Engenharia Elétrica

Les architectures de communication avionique actuelles impliquent un poids et des coûts d'intégration importants à cause de la quantité croissante du câblage et des connecteurs utilisés. Afin de répondre à ces besoins émergents, nous avons proposé dans cette thèse l'intégration des technologies sans fil dans le contexte avionique comme principale solution pour diminuer le poids et la complexité dus au câblage. Tout d’abord, nous avons conçu un réseau avionique de secours basé sur la technologie HRUWB, implémentant un protocole d'arbitrage TDMA et des divers mécanismes de fiabilité pour garantir les exigences de déterminisme et de sûreté. Par la suite, nous avons procédé à l'évaluation des performances de notre proposition en termes de délais en se basant sur des méthodes analytiques. Par ailleurs, nous avons étudié différentes solutions afin d’améliorer les marges d'évolutivité et de fiabilité du système. Enfin, nous avons validé notre réseau proposé à travers une étude de cas avionique réaliste; et les résultats obtenus ont mis en évidence la capacité de notre proposition à garantir les exigences du système en termes de déterminisme et de fiabilité. / The current avionics communication architecture inherits significant weight and integration costs due to the increasing quantity of wires and connectors. In addition, avionics interconnects are still subject to structural failure and fire hazard which decrease reliability and ramify the maintenance. To cope with these arising issues, integrating wireless technologies in avionics context is proposed in this thesis as a main solution to decrease the wiring-related weight and complexity.To achieve this aim, we design an alternative backup avionic network based on HR-UWB technology implementing a TDMA arbitration protocol and various reliability mechanisms to guarantee predictability and reliability requirements.Afterwards, we conduct analytical performance evaluation of our proposal in terms of delays. Moreover, we investigate different solutions to reach further enhancements on the system scalability and reliability.Finally, the validation of our proposal through a realistic avionics case study has been conduced, and the obtained results highlight its ability to guarantee the system requirements in terms of predictability and reliability.

本研究目的在建構個案學校翻轉教育的策略指標與探討其應用，除了提供個案學校對翻轉教育的回饋，亦提供有結構性的策略指標參照。研究方法上，先以文獻分析歸納整理相關指標，並與個案學校成員討論建構出學校翻轉教育之21項策略指標，接著以概念構圖法與集群分析整合學校成員對指標的重要性和執行率評估與分群以建立構面。最後求取各構面及指標權重，進行指標的重要-表現分析(IPA)，完成個案學校翻轉教育的策略指標體系。根據研究之結果與分析，歸納主要結論及建議如下： 一、結論 (一) 以概念構圖建構翻轉教育的4項構面與21項指標應用個案學校。 (二) 個案學校翻轉教育之重要性最高構面為「教師增能」。 (三) 個案學校翻轉教育之重要性最高的指標為「教師應有充足的備課時間」，其次為「提升教室的數位載具資源」、「教師所製作或選編之教學內容具吸引力」。 (四) 個案學校翻轉教育之執行表現最佳構面為「科技支援」；執行表現最差構面為「社群支持」。 (五) 個案學校翻轉教育之執行表現最佳指標為「提升教室的數位載具資源」，其次為「雲端課程平台分享」。 (六) 執行表現最差指標為「翻轉教學社群討論空間與時間安排」。 (七) IPA分析結果最急需改進的構面為「教師增能」。 (八) IPA分析結果最急需改進的指標為「教師應有充足的備課時間」，其次依序為「翻轉社群師資研習或學習機會」、「讓學生依照個別學習進度精熟學習」、「建立學生個人學習歷程檔案」、「教師能規劃翻轉學習之課程內容」。 (九) IPA分析的應用性。 二、建議 (一) 指標運用: 逕行利用指標繼續檢視翻轉教育歷程，利用指標繼續檢視個案學校往後的翻轉教育歷程。 (二) 應強化「教師增能」及「社群支持」。 (三) 應關注學生學習。 (四) 對未來研究之建議: 從個案研究的指標建構，推展至適用於全國性的樣本；持續針對指標系統做適當之更新。 / The purpose of this study is to construct the strategic indicators of school flipped education and to explore its application. In addition to providing individual case feedback on flip education, we also provide a structural reference to strategic indicators. As for research methods, by means of literature review, and discusses with the members of the case to construct 21 strategic indicators of school flipped education. Concept mapping questionnaire used to analyze the school members' opinion on the importance and implementation of each indicator and to help indicator selection. At last stage, find the weight of each indicator within each dimension and construct the system of indicators of school flipped education. The conclusions are follows: (1) This study constructs 4 main dimensions and 21 strategic indicators of school flipped education. (2) The highest aspect of the importance of school flipped education is " faculty enhancement ". (3) The most important indicator for flipped education is that "teachers should have sufficient preparation time" followed by "upgrading the classroom's digital vehicle resources" and "the teaching content produced or selected by teachers should be attractive." (4) The best performing aspect of the importance of school flipped education is "science and technology supporting"; the worst performing aspect is "community supporting". (5) The best indicator of the performance of school flipped education is "enhancement of digital mobile device in the Classroom" followed by "cloud-based course platform sharing". (6) The worst indicator of the performance of school flipped education is "flipped teaching community discussion space and schedule." (7) The most urgent aspect of IPA analysis is "faculty enhancement". (8) The most urgent indicator for IPA analysis is "teachers should have sufficient time for preparation", "community teachers' study or learning opportunities", "allowing students to master the learning process based on individual learning progress", " build student personal study files "," teachers can plan flip learning course content. " (9) The applicability of IPA analysis. According to the conclusions, some suggestions had been proposed: (1) Use of Indicators: Using indicators to continue to examine the course of overturning education, and using indicators to continue to examine cases of school rollover education in the future. (2) Teachers and community support should be strengthened. (3) Should pay attention to student learning. (4) Suggestions for future study: From the construction of indicators of case studies to the application of national samples; and continual updating of the indicator system.

This thesis contributes to the field of performance analysis in High Performance Computing with new concepts for in-memory event tracing. Event tracing records runtime events of an application and stores each with a precise time stamp and further relevant metrics. The high resolution and detailed information allows an in-depth analysis of the dynamic program behavior, interactions in parallel applications, and potential performance issues. For long-running and large-scale parallel applications, event-based tracing faces three challenges, yet unsolved: the number of resulting trace files limits scalability, the huge amounts of collected data overwhelm file systems and analysis capabilities, and the measurement bias, in particular, due to intermediate memory buffer flushes prevents a correct analysis. This thesis proposes concepts for an in-memory event tracing workflow. These concepts include new enhanced encoding techniques to increase memory efficiency and novel strategies for runtime event reduction to dynamically adapt trace size during runtime. An in-memory event tracing workflow based on these concepts meets all three challenges: First, it not only overcomes the scalability limitations due to the number of resulting trace files but eliminates the overhead of file system interaction altogether. Second, the enhanced encoding techniques and event reduction lead to remarkable smaller trace sizes. Finally, an in-memory event tracing workflow completely avoids intermediate memory buffer flushes, which minimizes measurement bias and allows a meaningful performance analysis. The concepts further include the Hierarchical Memory Buffer data structure, which incorporates a multi-dimensional, hierarchical ordering of events by common metrics, such as time stamp, calling context, event class, and function call duration. This hierarchical ordering allows a low-overhead event encoding, event reduction and event filtering, as well as new hierarchy-aided analysis requests. An experimental evaluation based on real-life applications and a detailed case study underline the capabilities of the concepts presented in this thesis. The new enhanced encoding techniques reduce memory allocation during runtime by a factor of 3.3 to 7.2, while at the same do not introduce any additional overhead. Furthermore, the combined concepts including the enhanced encoding techniques, event reduction, and a new filter based on function duration within the Hierarchical Memory Buffer remarkably reduce the resulting trace size up to three orders of magnitude and keep an entire measurement within a single fixed-size memory buffer, while still providing a coarse but meaningful analysis of the application. This thesis includes a discussion of the state-of-the-art and related work, a detailed presentation of the enhanced encoding techniques, the event reduction strategies, the Hierarchical Memory Buffer data structure, and a extensive experimental evaluation of all concepts.

Dans cette thèse, nous étudions les problèmes d'incertitudes dans les modèles probabilistes et tentons de déterminer leur impact sur l'analyse de performances et le dimensionnement des systèmes. Nous considérons deux aspects du problème d'imprécision. Le premier, consiste à étudier des chaînes en temps discret dont les probabilités ou taux de transition ne sont pas parfaitement connus. Nous construisons de nouveaux algorithmes de calcul de bornes par éléments sur les vecteurs de distribution stationnaires de chaînes partiellement spécifiées. Ces algorithmes permettent de déterminer des bornes par élément à chaque étape de calcul. Le second aspect étudié concerne le problème de mesures de traces de trafic réelles dans les réseaux. Souvent très volumineuses, la modélisation des traces de trafic est généralement impossible à effectuer de façon suffisamment précise et l'adéquation avec une loi de probabilité connue n'est pas assez réaliste. Utilisant une description par histogramme du trafic, nous proposons d'appliquer une nouvelle méthode d’évaluation de performance des réseaux. Fondée sur la comparaison stochastique pour construire des bornes optimales de supports réduits des histogrammes de trafics et sur la notion de monotonie stochastique des éléments de réseau, cette méthode permet de définir, de manière très pertinente, des garanties sur les mesures de performance. Nous obtenons en effet des bornes stochastiques supérieures et inférieures sur la longueur du tampon, les pertes, etc. L'intérêt et l'impact de notre méthode sont présentés sur diverses applications : éléments de réseau, AQM, réseaux de files d'attente, file avec processus d'arrivée non-stationnaire, etc / This thesis is devoted to the uncertainty in probabilistic models, how it impacts their analysis and how to apply these methods to performance analysis and network dimensioning. We consider two aspects of the uncertainty. The first consists to study a partially specified Markov chains. The missing of some transitions in the exact system because of its complexity can be solved by constructing bounding systems where worst-case transitions are defined to obtain an upper or a lower bound on the performance measures. We propose to develop new algorithms which give element-wise bounds of the steady-state distribution for the partially specified Markov chain. These algorithms are faster than the existing ones and allow us to compute element-wise bounds at each iteration.The second aspect studied concerns the problem of the measurements of real traffic trace in networks. Exact analysis of queueing networks under real traffic becomes quickly intractable due to the state explosion. Assuming the stationarity of flows, we propose to apply the stochastic comparison method to derive performance measure bounds under histogram-based traffics. We apply an algorithm based on dynamic programming to derive optimal bounding traffic histograms on reduced state spaces. Using the stochastic bound histograms and the monotonicity of the networking elements, we show how we can obtain, in a very efficient manner, guarantees on performance measures. We indeed obtain stochastic upper and lower bounds on buffer occupancy, losses, etc. The interest and the impact of our method are shown on various applications: elements of networks, AQM, queueing networks and queue with non-stationary arrival process

Cette thèse s'inscrit dans le cadre de la théorie des systèmes linéaires dans les dioïdes. Cette théorie concerne la sous-classe des systèmes à événements discrets modélisables par les Graphes d'Événements Temporisés (GET). La dynamique de ces graphes peut être représentée par des équations récurrentes linéaires sur des structures algébriques particulières telles que l'algèbre (max,+) ou l'algèbre (min,+).Ce mémoire est consacré à l'analyse de performances des systèmes dynamiques qui peuvent être modélisés graphiquement par des Graphes d'Événements Temporisés Généralisés (GETG). Ces derniers, contrairement au GET, n'admettent pas une représentation linéaire dans l'algèbre (min,+). Pour pallier à ce problème de non linéarité, nous avons utilisé une approche de modélisation définie sur un dioïde d'opérateurs muni de deux lois internes : loi additive correspondant à l'opération (min), et loi multiplicative équivalente à la loi de composition usuelle. Le modèle d'état obtenu, est utilisé pour évaluer les performances des GETG. Pour cela, nous avons proposé une nouvelle méthode qui a pour but de linéariser le modèle mathématique régissant l'évolution dynamique du modèle graphique, dans le but d'obtenir un modèle (min,+) linéaire. La deuxième partie de cette thèse est consacrée au problème qui consiste à déterminer les ressources à utiliser dans une ligne de production, en vue d'atteindre des performances souhaitée. Ceci est équivalent à déterminer le marquage initial de la partie commande du GETG. / This thesis is part of the theory of linear systems over dioids. This theory concerns the subclass of discrete event dynamic systems modeled by Timed Event Graphs (TEG). The dynamics of these graphs can be represented by linear recurrence equations over specific algebraic structures such as (max,+) algebra or (min,+) algebra.This report is devoted to the performance analysis of dynamic systems which can be represented graphically by Generalized Timed Event Graphs(GTEG). These type of graphs, unlike TEG, do not admit a linear representation in (min,+) algebra. To mitigate the problem of nonlinearity, we used a modeling approach defined on a dioid operators. The obtained state model is used to evaluate the performance of GTEG. For this, we proposed a new method to linearize the mathematical model governing the dynamic evolution of the graphical model in order to obtain a linear model in (min,+) algebra. The second part of this work is devoted to the problem of determining the resources to use in a production line, in order to achieve desired performance. These is equivalent to determining the initial marking of the control part of the GTEG.

O comportamento dinâmico das redes sem fio as torna muito peculiares e de difícil análise. No entanto, algumas destas redes, como as de sensores com funcionamento intermitente, redes periódicas ou cíclicas e as do sistema de satélites de órbita baixa têm um comportamento dinâmico relativamente previsível, pois as variações da topologia da rede no tempo são quase que determinísticas. Recentemente, um modelo teórico -- grafos evolutivos -- foi proposto com o intuito de capturar o comportamento dinâmico destas redes e formalizar algoritmos de roteamento de custo mínimo, além de outros. Os algoritmos e idéias obtidos com este modelo são teoricamente muito eficientes, mas, no entanto, antes deste trabalho não existiam estudos do uso destes modelos em situações práticas. Assim, o objetivo deste trabalho é analisar a aplicabilidade da teoria de grafos evolutivos na construção de protocolos de roteamento eficientes em cenários realistas. Foram implementados dois protocolos de roteamento para redes móveis ad hoc baseados nos algoritmos de grafos evolutivos, são eles: Jornada que Chega Mais Cedo e Jornada Mais Curta. Extensivas simulações foram realizadas utilizando o simulador de redes NS2 e os resultados foram comparados com outros quatro protocolos clássicos para este tipo de rede: AODV, DSR, OLSR e DSDV. Os resultados preliminares mostram que este recente modelo tem muito potencial para ser uma ferramenta poderosa no desenvolvimento e análise de algoritmos para redes dinâmicas com comportamento previsível. No entanto, foram apontados alguns aspectos que precisam ser melhores estudados para que estes algoritmos possam ser utilizados em situações reais. / The assessment of routing protocols for wireless networks is a difficult task, because of the networks\' highly dynamic behavior and the absence of benchmarks. However, some of these networks, such as intermittent wireless sensors networks, periodic or cyclic networks, and low earth orbit satellites systems, have more predictable dynamics, as the temporal variations in the network topology are somehow deterministic, which may make them easier to study. Recently, a graph theoretic model -- the evolving graphs -- was proposed to help to capture the dynamic behavior of these networks, in view of the construction of least cost routing and other algorithms. The algorithms and insights obtained through this model are theoretically very efficient and intriguing. However, before this work there was no study on the use of such theoretical results into practical situations. Therefore, the objective of our work is to analyze the applicability of the evolving graph theory in the construction of efficient routing protocols in realistic scenarios. We use the NS2 network simulator to first implement two evolving graph based routing protocols: Foremost Journey and Shortest Journey, They are evaluated and compared to four major ad-hoc protocols: AODV, DSR, OLSR and DSDV. Interestingly, our experiments show that evolving graphs have all the potentials to be an effective and powerful tool in the development and analysis of algorithms for dynamic networks, with predictable dynamics at least. In order to make this model widely applicable, however, some practical issues still have to be addressed and incorporated into the model.