La Sclérose Latérale Amyotrophique (SLA) est une affection neurodégénérative affectant sélectivement les motoneurones et conduisant au décès en 2 à 4 ans. Des facteurs génétiques, ainsi que diverses hypothèses physiopathologiques, telles que l’excitotoxicité et le stress oxydant, ont été évoqués pour expliquer la dégénérescence des motoneurones, mais aucune étiologie n’explique aujourd’hui la survenue de cette pathologie. Afin d’améliorer les connaissances des voies métaboliques impliquées dans la physiopathologie de la SLA, nous avons développé un modèle in vitro de co-Culture de motoneurones et d’astrocytes sur-Exprimant la Superoxyde Dismutase (SOD1) humaine sauvage ou mutée (SOD1G93C) et exposée au stress oxydant. Nous avons étudié les modifications de métabolisme après traitement oxydant par une approche métabolomique utilisant la chromatographie gazeuse couplée à la spectrométrie de masse et une analyse statistique multivariée des résultats. Ainsi nous avons observé une modification de métabolites impliqués notamment dans le cycle de Krebs, la neurotransmission excitatrice et la synthèse du glutathion, dans un modèle in vitro de SLA exposé au stress oxydant. / Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disorder affecting selectively motor neurons and leading to death in 2 to 4 years. Genetic factors and various pathophysiological hypotheses, such as excitotoxicity and oxidative stress, have been suggested to explain the degeneration of motor neurons, but today no etiology explains the occurrence of this disease. In order to improve the knowledge of the metabolic pathways involved in the pathogenesis of ALS, we developed an in vitro model of co-Culture of motor neurons and astrocytes over-Expressing human superoxide dismutase (SOD1), wild-Type or mutated (SOD1G93C), and exposed to oxidative stress. We studied the changes in metabolism after oxidative treatment with a metabolomics approach using gas chromatography-Mass spectrometry and multivariate statistical analysis. Thus we observed a change in metabolites involved in the citric acid cycle, the excitatory neurotransmission and the glutathione synthesis, in an in vitro model of ALS exposed to oxidative stress.

Critical networked services established between service provider and customers are expected to operate respecting Service Level Agreements (SLAs). An interesting possibility to monitor such SLAs is using active measurement mechanisms. However, these mechanisms are expensive in terms of network devices resource consumption and also increase the network load because of the injected traffic. In addition, if the number of SLA violations in a given time is higher than the number of available measurement sessions (common place in large and complex network infrastructures), certainly some violations will be missed. The current best practice, the observation of just a subset of network destinations driven by human administrators expertise, is error prone, does not scale well, and is ineffective on dynamic network conditions. This practice can lead to SLA violations being missed, which invariably affect the performance of several applications. In the present thesis, we advocated the use of Peer-to-Peer (P2P) technology to improve the detection of SLA violations. Such use is described using principles to control active measurement mechanisms. These principles are accomplished through strategies to activate measurement sessions. In this context, the major contributions of this thesis are: i) An approach to improve the detection of SLA violations through the steering of the activation of active measurement sessions using local and remote past service level measurement results and resource utilization constraints; ii) The concept of destination rank as an approach to autonomically prioritize destinations for the activation of active measurement sessions using destination scores; iii) The concept of correlated peers to enable the autonomic provisioning of a P2P measurement overlay for the exchange of relevant active measurement results; iv) The concept of virtual measurement sessions to enable the sharing of measurement results among correlated peers in order to save network devices resources and to improve SLA monitoring coverage; v) The definition of decentralized strategies to steer the activation of active measurement sessions using P2P principles. The method used on the investigation started with the execution of literature reviews on the networkwide control of measurement mechanisms and the employment of P2P technology on network management. After that, the proposed principles to control active measurement mechanisms and the strategies to activate measurement sessions were described. Finally, experiments were performed to evaluate the performance as well as to highlight properties of such principles and strategies. The findings showed properties which improve the detection of SLA violations in terms of the number of detected violations and the adaptivity to network dynamics. We expect that such findings can lead to better SLA monitoring tools and methods.

As technology advances so does the availability of computer-assisted learning software. Since the Swedish curriculum and syllabus in the subject English do not state how teachers should teach grammar or what grammatical items they should focus on, it is left to the teachers to decide themselves. This thesis aims to investigate how one can make use of CALL, Computer-Assisted Language Learning, when teaching grammar in the English language classroom. The following three research questions are posed:  • In what ways is CALL used in teaching grammar?  • Are any of the different ways of using CALL more successful than others?  • What are the students’ reactions to using CALL as a means of learning grammar?  The questions are answered by analyzing the theoretical background of second language acquisition, as well as by analyzing ten articles about learning grammar with the help of modern technology. The findings show that there are more ways of using technology in a teaching environment than there are articles about it, and this thesis only covers a few of the different means of using CALL to teach grammar. The findings also show that the results of computer-assisted teaching and learning are overall positive, but it cannot be concluded whether this is because of the software or the novelty of using CALL. Some software shows better results than other, such as error correction software. The overall perceptions of using CALL in the classroom are overwhelmingly positive from both teachers’ and students’ perspective. Furthermore, the results show that the students believe that using technology to learn helps them more than it actually does, showing how it helps motivate students to acquire new knowledge by making it more interesting for them. All in all, the findings of the research give teachers an overview of the current progress of CALL, as well as giving them suggestions about how to incorporate technology in their own teaching.

Processing instruction (PI) is a language teaching technique based upon the model of input processing developed by VanPatten (1993, 1996, 2002, 2004). The present study investigated the effects of PI as well as two other experimental conditions (traditional instruction and control) on the acquisition of the Spanish copulae ser and estar by 66 Chinese university students enrolled in a blended (partially presential and partially online) fourth-semester language course. The PI treatment condition included non-paradigmatic grammar explanations, processing strategies designed to help learners avoid commiting errors they may be predisposed to make, and structured input activities which eliminate redundant features of language that may make difficult the establishment of form-meaning connections. The traditional instruction (TI) treatment condition included paradigmatic explanations of grammar as well as production-based activities and exercises. Participants assigned to the control condition did not receive any instruction during the course of the study. The results of an experiment comparing the effects of each treatment condition on learners' scores on the Knowledge Test of Spanish Copulae, a measurement instrument designed for this study, found PI to be superior to TI at the immediate posttest level for tasks of interpretation and production. However, mean difference scores for the PI group were not significant when compared to those of the control group. Moreover, the learning gains exhibited by the PI group at immediate posttest were not durative, as they were not significant at the delayed posttest.

La sclérose latérale amyotrophique (SLA) et la démence fronto-temporale (DFT) sont deux maladies qui constituent un continuum clinico-pathologique. La mutation de FUS, une protéine nucléaire à fonctions multiples, provoque des cas familaux de SLA, et ces mutations provoquent une redistribution sub-cellulaire de FUS, du noyau vers le cytoplasme. Certains cas de DFT présentent une telles distribution anormale en l’absence de mutations de FUS. Il n’est pas connu si la maladie est provoquée par une perte de la fonction nucléaire de FUS et/ou un gain de fonction cytoplasmique.Nous avons généré et caractérisé une lignée de souris exprimant une forme cytoplasmique de FUS (Fus-ΔNLS). La localisation exclusive de FUS dans le cytoplasme provoque la mort des motoneurones via un gain de fonction dans les motoneurones eux-mêmes. Une localisation cytoplasmique partielle de FUS est suffisante pour développer un phénotype de la SLA et de DFT. Les mécanismes élucidés permettront de comprendre les bases des SLA/DFT. / Amyotrophic lateral sclerosis (ALS) and Frontotemporal dementia (FTLD) are now considered as a unique clinicopathological spectrum referred to as ALS/FTLD. Cytoplasmic aggregation of the physiologically nuclear FUS protein is a hallmark feature of a subset of ALS/FTLD. It remains unknonwn whether the critical pathogenic event relies on a loss of FUS normal nuclear functions, a toxic gain of function of FUS in the cytoplasm, or a combination of both.To answer this question we have generated a conditional mouse model expressing truncated FUS without nuclear localization signal - FusΔNLS. Our data showed that complete cytoplasmic mislocalization of truncated FUS protein within spinal motor neurons is a major determinant of motor neuron degeneration via toxic gain of function. A partial mislocalization of truncated FUS protein was sufficient to trigger key features of ALS and of FTLD.These studies allowed the elucidation of mechanisms underlying FUS role in ALS/FTLD, and will hopefully lead to development of therapies for these devastating diseases.

Avec l'avènement des technologies de Cloud computing et son adoption, les entreprises et les institutions académiques transfèrent de plus en plus leurs calculs et leurs données vers le Cloud. Alors que ce progrès et ce modèle simple d'accès ont eu un impact considérable sur notre communauté scientifique et industrielle en termes de réduction de la complexité et augmentation des revenus, les centres de données consomment énormément d'énergie, ce qui se traduit par des émissions plus élevées de C02. En réponse, de nombreux travaux de recherche se sont focalisés sur les enjeux du développement durable pour le Cloud à travers la réduction de la consommation d'énergie en concevant des stratégies d'efficacité énergétiques. Cependant, l'efficacité énergétique dans l'infrastructure du C!oud ne suffira pas à stimuler la réduction de • l'empreinte carbone. Il est donc impératif d'envisager une utilisation intelligente de l'énergie verte à la fois au niveau de l'infrastructure et de l'application pour réduire davantage l'empreinte carbone. Depuis peu, certains fournisseurs de Cloud computing alimentent leurs centres de données avec de l'énergie renouvelable. Les sources d'énergie renouvelable sont très intermittentes, ce qui crée plusieurs défis pour les gérer efficacement. Pour surmonter ces défis, nous étudions les options pour intégrer les différentes sources d'énergie renouvelable de manière réaliste et proposer un Cloud energy broker qui peut ajuster la disponibilité et la combinaison de prix pour acheter de l'énergie verte dynamiquement sur le marché de l'énergie et rendre les centres de données partiellement verts. Puis, nous introduisons le concept de la virtualisation de l'énergie verte, qui peut être vu comme une alternative au stockage d'énergie utilisé dans les centres de données pour éliminer le problème d'intermittence dans une certaine mesure. Avec l'adoption du concept de virtualisation, nous maximisons l'utilisation de l'énergie verte contrairement au stockage d'énergie qui induit des pertes d'énergie, tout en introduisant des Green SLA basé sur l'énergie verte pour le fournisseur de services et les utilisateurs finaux. En utilisant des traces réalistes et une simulation et une analyse approfondie, nous montrons que la proposition peut fournir un système efficace, robuste et rentable de gestion de l'énergie pour le centre de données. Si une gestion efficace de l'énergie en présence d'énergie verte intermittente est nécessaire, la façon dont les applications Cloud modernes peuvent tirer profit de la présence ou l'absence d'énergie verte n'a pas été suffisamment étudiée. Contrairement aux applications Batch, les applications Interactive Cloud doivent toujours être accessibles et ne peuvent pas être programmées à l'avance pour correspondre au profil d'énergie verte. Par conséquent, cette thèse propose une solution d'autoscaling adaptée à l'énergie pour exploiter les caractéristiques internes des applications et créer une conscience d'énergie verte dans l'application, tout en respectant les propriétés traditionnelles de QoS. Pour cela, nous concevons un contrôleur d'application green qui profite de la disponibilité de l'énergie verte pour effectuer une adaptation opportuniste dans une application gérée par un contrôleur orienté performance. L'expérience est réalisée avec une application réelle sur Grid5000 et les résultats montrent une réduction significative de la consommation d'énergie par rapport à l'approche orientée performance, tout en respectant les attributs traditionnels de QoS. / With the advent of cloud enabling technologies and adoption of cloud computing, enterprise and academic institutions are moving their IT workload to the cloud. Although this prolific advancement and easy to access model have greatly impacted our scientific and industrial community in terms of reducing complexity and increasing revenue, data centers are consuming enormous amount of energy, which translates into higher carbon emission. In response, varieties of research work have focused on environmental sustainability for Cloud Computing paradigm through energy consumption reduction by devising energy efficient strategies. However, energy efficiency in cloud infrastructure alone is not going to be enough to boost carbon footprint reduction. Therefore, it is imperative to envision of smartly using green energy at infrastructure and application level for further reduction of carbon footprint. In recent years, some cloud providers are powering their data centers with renewable energy. The characteristics of renewable energy sources are highly intermittent which creates several challenges to manage them efficiently. To overcome the problem, we investigate the options and challenges to integrate different renewable energy sources in a realistic way and propose a Cloud energy broker, which can adjust the availability and price combination to buy Green energy dynamically from the energy market in advance to make a data center partially green. Later, we introduce the concept of Virtualization of Green Energy, which can be seen as an alternative to energy storage used in data center to eliminate the intermittency problem to some extent. With the adoption of virtualization concept, we maximize the usage of green energy contrary to energy storage which induces energy losses, while introduce Green Service Level Agreement based on green energy for service provider and end users. By •using realistic traces and extensive simulation and analysis, we show that, the proposal can provide an efficient, robust and cost-effective energy management scheme for data center. While an efficient energy management in the presence of intermittent green energy is necessary, how modern Cloud applications can take advantage of the presence/absence of green energy has not been studied with requisite effort. Unlike Batch applications, Interactive Cloud applications have to be always accessible and car not be scheduled in advance to match with green energy profile. Therefore, this thesis proposes an energy adaptive autoscaling solution to exploit applications internal to create green energy awareness in the application, while respecting traditional QoS properties. To elaborate, we design green energy aware application controller that takes advantage of green energy availability to perform opportunistic adaptation in an application along with performance aware application controller. Experiment is performed with real life application at Grid5000 and results show significant reduction of energy consumption while respecting traditional QoS attributes compared to performance aware approach.

Do adult L2 English bilingual speakers have difficulty with cognate words whose meanings are distinct across their two languages? This study explored the extent to which variations in meaning in cross-language cognates affect translation performance in a translation task by L2 English (L1 Spanish) speakers who learned English as adults. A prep-phase experiment was conducted to test native English-speakers’ predicted completions of the study’s stimuli sentences, in order to choose the optimal stimuli for the primary experiment. The method for the primary experiment of this study consisted of a web-based translation task of 120 sentences from Spanish to English, while controlling for polysemy and frequency. The results showed that adult L2 learners of English did experience difficulty when translating cognates in sentences from their L1 to their L2. The interaction of the Spanish word’s polysemous nature, Spanish word frequency, English target frequency and English cognate frequency played a role in the participants’ performance.

The thesis concerns with measurement of processes of support and maintenance of a specialized information system for petrol stations. It is a system for support of retail sale of fuel and other non-fuel goods. As the main objective of the thesis was selected design of set of metrics for measurement of these processes based on multidimensionality principle. It means that dimensions and indicators are designed separately and a set of recommended metrics is based on their combinations. It is possible to get additional metrics using other combinations. The theoretical basis of the thesis is the ITIL library -- which is an internationally respected standard for IT service management -- and metrics designed in other literature and diploma theses. The designed metrics are adapted for the analyzed environment and service information system, however, they should be usable in other environments as well. Selected metrics were implemented to the service information system and measured in a relatively wide sample of input data. This allowed evaluating usability of the metrics and comparison of the threshold values with the measured ones. The relationship of metrics and service agreements (SLAs) is described in the end of the thesis. Specific metrics suitable to be incorporated to SLA were designed including a real-life example.

The thesis is focused on administration and supervision of Service oriented architecture (SOA). The main idea of SOA is system decomposition to functional units, which contain bounded and well understandable functionality. This functionality, in form of service, is provided to other units through clearly described interface. Well-identified relationships between service provider and service consumer are created. SOA governance aims to manage these relationships, to monitor quality of services and to control adherence of stated rules and policies. These rules and policies are established by central team responsible for integration and architecture development. Starting point of successful architecture development is acquaintance of service portfolio and their inter-relationships. Each service should go through whole formal life-cycle, which particular phases have special policies applied. Main interest is devoted to run-time phase - run-time monitoring and run-time governance. Theory is complemented by practical examples of report generation and service monitoring with SLA adherence observation. The thesis also contains structural model of SOA governance taking into account infrastructural elements of SOA and information flows between them.

The diploma thesis deals with the theme of Modeling business processes with orientation to the metric process interface. The aim of the thesis is to design the architecture of metric, which enabled to construct type metrics of the process interface. Firstly, the reader is informed about the basic terms related to process modeling, process management and connection of strategy with metrics. The next part describes the architecture of metrics created, which is oriented to supporting the process management of the organization. The set of type metrics was chosen with consideration to the possibility of their placement on process interface. The set created in this manner should help to connect strategy with processes when implementing into organization. The last part of the thesis describes the practical application of knowledge obtained in the PARMA project. This project deals with the proposal and implementation of the process management into the Bureau of the Municipality of Prague 10. The thesis deals with two processes and the conclusions obtained in the theoretic part of the thesis are applied herein.