Submitted by Cláudia Bueno (claudiamoura18@gmail.com) on 2016-03-03T19:14:52Z No. of bitstreams: 2 Dissertação - Cleon Xavier Pereira Júnior - 2015.pdf: 1262859 bytes, checksum: 35b1053bbf6dfda4e73013526fab8f6b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-03-04T11:15:57Z (GMT) No. of bitstreams: 2 Dissertação - Cleon Xavier Pereira Júnior - 2015.pdf: 1262859 bytes, checksum: 35b1053bbf6dfda4e73013526fab8f6b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2016-03-04T11:15:57Z (GMT). No. of bitstreams: 2 Dissertação - Cleon Xavier Pereira Júnior - 2015.pdf: 1262859 bytes, checksum: 35b1053bbf6dfda4e73013526fab8f6b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2015-10-13 / Digital ink technology is available in several electronic devices and can bring contributions to the evaluation process as it allows access to information that was not available in assessments carried out using traditional methods. It is particularly interesting in evaluations involving drawings, since it offers an environment quite similar to pencil and paper, with the advantage of allowing process automation. Thus, the first step is to provide resources to capture, store and reproduce the design done. Then, it is necessary to analyze the collected data. However, analysis of drawings is a complex task and focus of several research projects. Thus, as an initial contribution, this work focuses on the analysis of simple traces. In order to explore digital ink as a means of evaluation, a tool was developed to automate the capture, storage and reproduction of drawings, and test the features and limitations of the analysis of digital ink in order to extract relevant knowledge for the evaluator to make decisions. To test the tool, psychological tests were selected, since in this knowledge domain the use of drawings for assessment is widespread. The tool was developed in two stages. The first, more general, offers resources for test application, storage and later playback for an analysis by the evaluator, not providing resources for automatic analysis of results. Though simple, these resources offer a great contribution, because in addition to storing the final result, which can be visualized as an image, it also stores the design process (allowing the evaluator to follow step by step how the test was performed) in addition to storing other information such as time spent to perform certain steps, the use of rubber etc. The second stage of the tool is the analysis of the stored data. This analysis is test dependent, and should be implemented according to it’s evaluation requirements. As a case study, a test, that uses simple traces, was implemented. The result showed that the digital ink has advantages for carrying out evaluations using drawings as a medium, and should be subject to further research aimed at automating tests with more complex drawings. / A tinta digital está presente em vários dispositivos eletrônicos e pode trazer contribuições para o processo de avaliação por permitir acesso a informações que não estavam disponíveis quando realizadas usando métodos tradicionais. Ela é particularmente interessante em avaliações que envolvem desenhos, já que oferece um meio bastante similar ao lápis e papel com a vantagem de permitir a automatização do processo. Sendo assim, o primeiro passo é oferecer recursos para capturar, armazenar e reproduzir o desenho realizado. Em seguida, é necessário analisar os dados coletados. No entanto, a análise de desenhos é uma tarefa complexa e alvo de diversas pesquisas. Assim, como uma contribuição inicial, este trabalho foca a análise de traços simples (retas). Com o objetivo de explorar a tinta digital como meio de avaliação foi desenvolvida uma ferramenta para automatizar a captura, armazenamento e reprodução de desenhos, e testar as funcionalidades e limitações da análise da tinta digital visando extrair conhecimento relevante para o avaliador poder tomar decisões. Para testar a ferramenta foram selecionados testes psicológicos, já que neste domínio de conhecimento o uso do desenho é bastante difundido. A ferramenta foi desenvolvida em duas etapas. A primeira, mais geral, oferece recursos para a aplicação do teste, seu armazenamento e posterior reprodução para uma análise feita pelo avaliador, não oferecendo recursos para a análise automática dos resultados. Apesar de simples, estes recursos já oferecem uma grande contribuição, pois além de armazenar como imagem o resultado final, armazena também o processo do desenho (permitindo que o avaliador acompanhe passo a passo como o teste foi realizado) e informações adicionais como tempo gasto para atividades, uso da borracha, etc. A segunda etapa da ferramenta é a análise dos dados armazenados. Esta análise é dependente do teste, e deve ser implementada de acordo com os requisitos da avaliação. Como estudo de caso foi implementado um teste que usa traços simples. O resultado mostrou que a tinta digital apresenta vantagens para a realização de avaliações que usam desenhos como meio, e deve ser alvo de futuras pesquisas visando a automatização de testes com desenhos mais complexos.

L'auteur n'a pas fourni de résumé en français / High Performance Computing is preparing the era of the transition from Petascale to Exascale. Distributed computing systems are already facing new scalability problems due to the increasing number of computing resources to manage. It is now necessary to study in depth these systems and comprehend their behaviors, strengths and weaknesses to better build the next generation.The complexity of managing users applications on the resources conducted to the analysis of the workload the platform has to support, this to provide them an efficient service.The need for workload comprehension has lead to the collection of traces from production systems and to the proposal of a standard workload format. These contributions enabled the study of numerous of these traces. This also lead to the construction of several models, based on the statistical analysis of the different workloads from the collection.Until recently, existing workload traces did not enabled researchers to study the consumption of resources by the jobs in a temporal way. This is now changing with the need for characterization of jobs consumption patterns.In the first part of this thesis we propose a study of existing workload traces. Then we contribute with an observation of cluster workloads with the consideration of the jobs resource consumptions over time. This highlights specific and unattended patterns in the usage of resources from users.Finally, we propose an extension of the former standard workload format that enables to add such temporal consumptions without loosing the benefit of the existing works.Experimental approaches based on workload models have also served the goal of distributed systems evaluation. Existing models describe the average behavior of observed systems.However, although the study of average behaviors is essential for the understanding of distributed systems, the study of critical cases and particular scenarios is also necessary. This study would give a more complete view and understanding of the performance of the resources and jobs management. In the second part of this thesis we propose an experimental method for performance evaluation of distributed systems based on the replay of production workload trace extracts. These extracts, replaced in their original context, enable to experiment the change of configuration of the system in an online workload and observe the different configurations results. Our technical contribution in this experimental approach is twofold. We propose a first tool to construct the environment in which the experimentation will take place, then we propose a second set of tools that automatize the experiment setup and that replay the trace extract within its original context.Finally, these contributions conducted together, enable to gain a better knowledge of HPC platforms. As future works, the approach proposed in this thesis will serve as a basis to further study larger infrastructures.

Le travail doctoral concerne l’étude des dynamiques de changement en faveur de pratiques éco-citoyennes et liées à la mobilité via la communication numérique engageante. La question de recherche qui en découle relève de la définition des conditions et des modalités portant sur la possibilité de transposer les travaux fondateurs antérieurs consacrés à la communication engageante dans le cadre élargi d’interactions médiatées et médiatisées sur l’internet, en milieu naturel.Pour tenter de répondre à cette question, nous avons situé notre propos dans un cadre théorique et méthodologique ouvert en montrant les relations entre la communication engageante et d’autres perspectives relevant du lien, du rite d’interaction ; nous abordons également la problématisation du changement individuel et collectif. Nos travaux apportent une contribution à la réflexion concernant la relation entre information et communication. Sur le plan empirique, nous avons mené une recherche-action dans une Université du Sud de la France. Il s’agissait, dans le contexte d’un partenariat avec cette institution, d’amener des étudiants à s’inscrire sur un dispositif Web de mise en relation de covoitureurs potentiels. Nous avons mené deux expériences mobilisant le paradigme de la communication numérique engageante.Notre approche, replacée dans les problématiques et la méthode de la recherche-action, n’est pas exempte de limites, notamment liées à une validité interne modérée par certains éléments hors de notre contrôle : facteurs historiques, mortalité expérimentale, variables parasites…Tous ces éléments sont présentés, discutés, mis en perspective. / This doctoral work relates to the study of change dynamics in favor of eco-friendly practices and tied to mobility through digital binding communication. The research question that ensues concerns the definition of the terms and conditions about the possibility of incorporating previous work on digital binding communication in the broader context of mediated and mediatized interactions on the internet, in a natural environment.In order to try answering this question, we set our discussion on an open theoretical and methodological framework by showing the relationships between binding communication and other perspectives concerning links, interaction rituals, we also broach upon the problematization of individual and collective change. Our work provides a contribution to the debate on the relationship between information and communication. From an empirical standpoint, we conducted an action research at a southern French university. Our goal was within the context of a partnership with this institution, to encourage students to register to a Web device putting potential carpoolers in touch. We conducted two experiments mobilizing the digital binding communication paradigm.Our approach, put in the context of the set of problems and the action research method is not without flaws, especially insofar as its internal validity is weakened by elements beyond our control: historical factors, experimental mortality, confounding variables…All these elements are presented, discussed, put into perspective.