Le présent travail est une étude sur Le Petit Prince, d’Antoine de Saint-Exupéry. Le but principal du mémoire est d’analyser la réception du message chez le lecteur adulte et chez le lecteur enfant. Nous avons analysé comment l’enfant et l’adulte interprètent le message donné par l’oeuvre, parsemée d’images et d’allégories, des citations qui son devenues symboliques avec le temps. Ensuite, nous avons analysé les personnages dans le but d’interpréter leur sens symbolique. Pour réaliser cette recherche, nous avons regardé plusieurs interprétations allégoriques. Les résultats de notre étude nous ont permis d’affirmer que la réception du message n’est pas la même chez le lecteur adulte et le lecteur enfant. Les résultats de cette étude montrent enfin qu’il est trop difficile pour le lecteur enfant de comprendre le sens moral d’une image allégorique, tandis que le lecteur adulte est plus apte à interpréter l’image allégorique en s’appuyant sur son expérience de la lecture et de son expérience de la vie quotidienne. L’auteur utilise l’image symbolique dans la littérature pour montrer avec plus de clarté la condition humaine. En conclusion nous apprenons que Le Petit Prince nous fait réfléchir sur la vie de la façon plus profonde et philosophique. / The current work consists of a study of The Little Prince, written by Antoine de Saint-Exupéry. The purpose of this study is to analyze the reception of the message given to the adult reader and the child reader. We have analyzed how the child and the adult interpret the work containing images and allegories. Then we have analyzed the fictional characters in order to interpret their symbolic meaning. To conduct this research we have looked at various allegorical interpretations. The result of our study shows that the reception of the message varies between the adult reader and the child reader. The result of this study also shows that it is too 3 difficult for the child reader to understand the moral sense of an allegorical image, while the adult reader is able to interpret the allegorical picture more easily by relying on the reading and life experiences. The author uses the symbolic image of the literature to clarify the truth about man and human society. In conclusion we learn that the work of the Little Prince makes us reflect upon life with a deeper philosophical meaning.

Dans cette thèse, nous étudions le problème d’optimisation du temps de transfert de données dans les systèmes de gestion de données distribuées, en nous focalisant sur la relation entre le temps de communication de données et la configuration du middleware. En réalité, le middleware détermine, entre autres, comment les données sont divisées en lots de F tuples et messages de M octets avant d’être communiqués à travers le réseau. Concrètement, nous nous concentrons sur la question de recherche suivante : étant donnée requête Q et l’environnement réseau, quelle est la meilleure configuration de F et M qui minimisent le temps de communication du résultat de la requête à travers le réseau?A notre connaissance, ce problème n’a jamais été étudié par la communauté de recherche en base de données.Premièrement, nous présentons une étude expérimentale qui met en évidence l’impact de la configuration du middleware sur le temps de transfert de données. Nous explorons deux paramètres du middleware que nous avons empiriquement identifiés comme ayant une influence importante sur le temps de transfert de données: (i) la taille du lot F (c’est-à-dire le nombre de tuples dans un lot qui est communiqué à la fois vers une application consommant des données) et (ii) la taille du message M (c’est-à-dire la taille en octets du tampon du middleware qui correspond à la quantité de données à transférer à partir du middleware vers la couche réseau). Ensuite, nous décrivons un modèle de coût permettant d’estimer le temps de transfert de données. Ce modèle de coût est basé sur la manière dont les données sont transférées entre les noeuds de traitement de données. Notre modèle de coût est basé sur deux observations cruciales: (i) les lots et les messages de données sont communiqués différemment sur le réseau : les lots sont communiqués de façon synchrone et les messages dans un lot sont communiqués en pipeline (asynchrone) et (ii) en raison de la latence réseau, le coût de transfert du premier message d’un lot est plus élevé que le coût de transfert des autres messages du même lot. Nous proposons une stratégie pour calibrer les poids du premier et non premier messages dans un lot. Ces poids sont des paramètres dépendant de l’environnement réseau et sont utilisés par la fonction d’estimation du temps de communication de données. Enfin, nous développons un algorithme d’optimisation permettant de calculer les valeurs des paramètres F et M qui fournissent un bon compromis entre un temps optimisé de communication de données et une consommation minimale de ressources. L’approche proposée dans cette thèse a été validée expérimentalement en utilisant des données issues d’une application en Astronomie. / In this thesis, we take a complementary look to the problem of optimizing the time for communicating query results in distributed query processing, by investigating the relationship between the communication time and the middleware configuration. Indeed, the middleware determines, among others, how data is divided into batches and messages before being communicated over the network. Concretely, we focus on the research question: given a query Q and a network environment, what is the best middleware configuration that minimizes the time for transferring the query result over the network? To the best of our knowledge, the database research community does not have well-established strategies for middleware tuning. We present first an intensive experimental study that emphasizes the crucial impact of middleware configuration on the time for communicating query results. We focus on two middleware parameters that we empirically identified as having an important influence on the communication time: (i) the fetch size F (i.e., the number of tuples in a batch that is communicated at once to an application consuming the data) and (ii) the message size M (i.e., the size in bytes of the middleware buffer, which corresponds to the amount of data that can be communicated at once from the middleware to the network layer; a batch of F tuples can be communicated via one or several messages of M bytes). Then, we describe a cost model for estimating the communication time, which is based on how data is communicated between computation nodes. Precisely, our cost model is based on two crucial observations: (i) batches and messages are communicated differently over the network: batches are communicated synchronously, whereas messages in a batch are communicated in pipeline (asynchronously), and (ii) due to network latency, it is more expensive to communicate the first message in a batch compared to any other message that is not the first in its batch. We propose an effective strategy for calibrating the network-dependent parameters of the communication time estimation function i.e, the costs of first message and non first message in their batch. Finally, we develop an optimization algorithm to effectively compute the values of the middleware parameters F and M that minimize the communication time. The proposed algorithm allows to quickly find (in small fraction of a second) the values of the middleware parameters F and M that translate a good trade-off between low resource consumption and low communication time. The proposed approach has been evaluated using a dataset issued from application in Astronomy.

Enterprise Systems like customer-billing systems or financial transaction systems are required to process large volumes of data in a fixed period of time. Those systems are increasingly required to also provide near-time processing of data to support new service offerings. Common systems for data processing are either optimized for high maximum throughput or low latency. This thesis proposes the concept for an adaptive middleware, which is a new approach for designing systems for bulk data processing. The adaptive middleware is able to adapt its processing type fluently between batch processing and single-event processing. By using message aggregation, message routing and a closed feedback-loop to adjust the data granularity at runtime, the system is able to minimize the end-to-end latency for different load scenarios. The relationship of end-to-end latency and throughput of batch and message-based systems is formally analyzed and a performance evaluation of both processing types has been conducted. Additionally, the impact of message aggregation on throughput and latency is investigated. The proposed middleware concept has been implemented with a research prototype and has been evaluated. The results of the evaluation show that the concept is viable and is able to optimize the end-to-end latency of a system. The design, implementation and operation of an adaptive system for bulk data processing differs from common approaches to implement enterprise systems. A conceptual framework has been development to guide the development process of how to build an adaptive software for bulk data processing. It defines the needed roles and their skills, the necessary tasks and their relationship, artifacts that are created and required by different tasks, the tools that are needed to process the tasks and the processes, which describe the order of tasks.

Dans le contexte général du contrôle du poids, le développement de produits à haut pouvoir satiétogène apparaît comme une alternative intéressante ou une réponse complémentaire aux propositions de régimes réduits en calories. Le principal objectif de cette thèse est d'évaluer le pouvoir satiétogène simple et combinés des fibres et des protéines dans une matrice biscuitière ainsi que l'influence de facteurs psychologiques telle que l'information fournie sur un aliment sur la satiété. Concernant l'influence de paramètres psychologiques, si l'on s'en réfère aux limites du TFEQ couramment utilisées par Stunkard et Messick, une cohorte de jeunes adultes français a de faibles score de TFEQ pour la restriction cognitive, la désinhibition et la faim. Ce résultat nous a permis de proposer de nouvelles limites plus basses que celles couramment utilisées et adaptées à cette catégorie de la population, souvent étudiée dans les études cliniques de comportement alimentaire. Ces limites ont été utilisées pour définir nos groupes de volontaires pour l'étude sur l'effet d'un message sur la satiété. Nous avons conclu qu'un message sur les propriétés satiétogènes d'un produit pourrait influencer un sous-groupe de la population, que sont les femmes faiblement restreintes et fortement désinhibées. Afin d'optimiser la sélection des prototypes utilisés dans l'étude clinique sur l'effet des fibres et des protéines sur la satiété, nous avons développé avec succès un nouvel outil pour entraîner un panel d'experts sensoriels à évaluer leur satiété. Un mélange de fibres et de protéines réduit les sensations d'appétit de manière plus importante que les biscuits enrichis simplement en un des ingrédients. Cette combinaison d'ingrédients actifs stimule des mécanismes qui sont induits par les fibres (comme la réduction de l'insulinémie et le ralentissement de la vidange gastrique) et les protéines (comme la diminution de la glycémie). En conclusion, cette approche intégrative de la satiété a permis d'apporter de nouvelles connaissances sur les effets synergique des fibres et des protéines sur les sensations d'appétit. Nous avons aussi proposé de nouvelles méthodes pour évaluer le pouvoir satiétogène de différents produits en utilisant l'expertise des domaines des sciences du sensoriel et de la nutrition. Ce travail souligne aussi l'importance des aspects psychologiques dans les études de nutrition humaine.Mots clés: Satiété, Fibres, Protéines, Three-Factor Eating questionnaire, Message satiété, entraînement à l'évaluation de la satiété / In the general context of weight management, the development of food with higher satiating power appears to be an interesting alternative or a complementary answer to the proposition of calorie-reduced diets. The main aim of this thesis was to evaluate the single and combined satiating effects of proteins and fibres in biscuits and the influence of psychological factors, such as a given message about a food, on satiety. As for cognitive parameters, considering Stunkard and Messick guidelines about TFEQ limits, young French adults aged 20–39 yrs have low cognitive restraint, disinhibition and hunger scores. This finding makes it possible to suggest new limits for the TFEQ scores better adapted to this category of the population which is often used in clinical studies. We used these limits to define our groups of volunteers in the study about the influence of a message on satiety. It was found that giving a message on the satiating potential of a food may affect a sub-group of the population, such as low-restrained high-disinhibited women. To optimize the selection of the prototypes used in the clinical study about fibre and protein effects on satiety, we developed a new successful methodology to train a panel of sensorial experts in satiety assessment. A combination of fibre and protein in biscuits reduces appetite sensations more than do single enrichment in fibre or protein. The mixing of fibre and protein activates mechanisms which are induced by fibre (such as reduced insulinemia and slowed gastric emptying) and protein (such as reduced glycaemia). In conclusion, this integrative approach on satiety brings new knowledge about the synergistic effect of fibre and protein on appetite sensations. It proposes a new methodology using sensory and nutrition expertise to assess the satiety power of products and it also highlights the importance of psychometric traits in human food studies.Key words: Satiety, Fibres, Proteins, Three-Factor Eating questionnaire, Satiety message, training in satiety assessment

We study how to perform effective Bayesian inference in high-dimensional sparse Factor Analysis models with a zero-norm, sparsity-inducing prior on the model parameters. Such priors represent a methodological ideal, but Bayesian inference in such models is usually regarded as impractical. We test this view. After empirically characterising the properties of existing algorithmic approaches, we use techniques from statistical mechanics to derive a theory of optimal learning in the restricted setting of sparse PCA with a single factor. Finally, we describe a novel `Dense Message Passing' algorithm (DMP) which achieves near-optimal performance on synthetic data generated from this model.DMP exploits properties of high-dimensional problems to operate successfully on a densely connected graphical model. Similar algorithms have been developed in the statistical physics community and previously applied to inference problems in coding and sparse classification. We demonstrate that DMP out-performs both a newly proposed variational hybrid algorithm and two other recently published algorithms (SPCA and emPCA) on synthetic data while it explains at least the same amount of variance, for a given level of sparsity, in two gene expression datasets used in previous studies of sparse PCA.A significant potential advantage of DMP is that it provides an estimate of the marginal likelihood which can be used for hyperparameter optimisation. We show that, for the single factor case, this estimate exhibits good qualitative agreement both with theoretical predictions and with the hyperparameter posterior inferred by a collapsed Gibbs sampler. Preliminary work on an extension to inference of multiple factors indicates its potential for selecting an optimal model from amongst candidates which differ both in numbers of factors and their levels of sparsity.

Magister Scientiae - MSc / In South Africa, Deaf people communicate with one another and the broader community by means of South African Sign Language. The majority of Deaf people who have access to a mobile phone (cell phone) use Short Message Service (SMS) to communicate and share information with hearing people, but seldom use it among themselves. It is assumed that video messaging will be more accessible to Deaf people, since their level of literacy may prevent them from making effective use of information that is disseminated via texting/SMS. The principal objective of the esearch was to explore a cost-effective and efficient mass multimedia messaging system. The intention was to adapt a successful text-based mass notification system, developed by a local nongovernmental organization (NGO), to accommodate efficient and affordable video mass messaging for Deaf people. The questions that underpin this research are: How should video- streaming mass-messaging methods be compared and evaluated to find the most suitable method to deliver an affordable and acceptable service to Deaf people? What transport vehicles should be considered: Multimedia Message Service (MMS), the web, electronic mail, or a cell phone resident push/pullapplication? Which is the most cost effective? And, finally: How does the video quality of the various transport vehicles differ in terms of the clarity of the sign language as perceived by the Deaf? The soft-systems methodology and a mixed-methods methodology were used to address the research questions. The soft-systems methodology was followed to manage the research process and the mixed-methods research methodology was followed to collect data. Data was collected by means of experiments and semi-structured interviews. A prototype for mobile phone usage was developed and evaluated with Deaf members the NGO Deaf Community of Cape Town. The technology and internet usage of the Deaf participants provided background information. The Statistical Package for Social Science (SPSS) was used to analyse the quantitative data, and content analysis was used to analyse the documents and interviews. All of the Deaf participants used their mobile phones for SMS and the majority (81.25%) used English to type messages; however, all indicated that they would have preferred to use South Africa sign language on their mobile phones if it were available. And they were quite willing to pay between 75c and 80c per message for using such a video-messaging service.Of the transport vehicles demonstrated, most Deaf people indic indicated that they preferred to use the SMS prototype (with a web link to the video) rather than the MMS prototype with the video attached. They were, however, very concerned about the cost of using the system, as well as the quality of the sign language videos. / South Africa

O modelo de atores tem sido visto como uma abordagem alternativa à programação concorrente convencional, baseada em travas e variáveis de condição. Atores são agentes computacionais que se comunicam por troca de mensagens e que possuem uma caixa de correio e um comportamento. As mensagens destinadas a um ator são armazenadas na caixa de correio do ator e processadas de maneira assíncrona. Sistemas de middleware orientados a mensagens trabalham com troca assíncrona de mensagens e formam uma base que simplifica o desenvolvimento de aplicações distribuídas. Tais sistemas permitem interoperabilidade com baixo acoplamento e provêm suporte para tratamento robusto de erros em caso de falhas. Message brokers são frequentemente apresentados como uma tecnologia que pode mudar a maneira com que sistemas distribuídos são construídos. A especificação AMQP é uma proposta recente de padronização de um protocolo para message brokers. Neste trabalho exploramos a potencial sinergia entre um message broker e uma implementação do modelo de atores. Criamos uma versão modificada da implementação do modelo de atores do projeto Akka que utiliza um message broker AMQP como mecanismo de transporte de mensagens para atores remotos. / The actor model has been seen as an alternative for conventional concurrent programming based on locks and condition variables. Actors are computational agents that communicate by sending messages and have a mailbox and a behavior. The messages sent to an actor are stored in its mailbox and are asynchronously processed. Message oriented middleware systems work with asynchronous message exchange and create a base that simplifies the development of distributed applications. These systems have interoperability with low coupling and provide support for robust error handling in case of failures. Message brokers are often presented as a technology that can change the way distributed systems are built. The AMQP specification is a recent proposal of a standard protocol for message brokers. In this document we explore the potential synergy between a message broker and an implementation of the actor model. We created a modified version of the actor model implementation provided by the Akka project. Our modified implementation uses an AMQP message broker as the transport engine for messages to remote actors.

This master thesis focuses on problematics of GOOSE messages used for evaluation of blocking conditions of substations in the electrical power plant in Opatovice (EOP). Firstly, the thesis describes the multifunctional electrical protection relay 7SJ64 as well as GOOSE messages, that are run by protocol IEC 61850. Next, it contains a brief description of the energy consumption of EOP itself, since the project was written for its protection relays. In the practical part of the thesis, an experimental network consisting of two distinct substations with three 7SJ64 multifunctional protection relays was designed together with simple blocking conditions. These require information about the state of circuit breakers. The whole communication is carried by GOOSE messages. Moreover, to inform about the loss of GOOSE messages, the logical function SI_GET_S (decoder) was used inside the coding interface of DIGSI programme called CFC. This function was then tested in the experimental network using each different condition, that experienced interruptions of communication caused either by simulated defect of optical cabels, or by „user‘s mistakes“ in the settings of communication parameters in DIGSI. Finally, the system protecting from the loss of GOOSE messages was incorporated into the existing project evaluating blocking conditions in the block substation of EOP.

Le mode adjoint de la Différentiation Algorithmique (DA) est particulièrement intéressant pour le calcul des gradients. Cependant, ce mode utilise les valeurs intermédiaires de la simulation d'origine dans l'ordre inverse à un coût qui augmente avec la longueur de la simulation. La DA cherche des stratégies pour réduire ce coût, par exemple en profitant de la structure du programme donné. Dans ce travail, nous considérons d'une part le cas des boucles à point-fixe pour lesquels plusieurs auteurs ont proposé des stratégies adjointes adaptées. Parmi ces stratégies, nous choisissons celle de B. Christianson. Nous spécifions la méthode choisie et nous décrivons la manière dont nous l'avons implémentée dans l'outil de DA Tapenade. Les expériences sur une application de taille moyenne montrent une réduction importante de la consommation de mémoire. D'autre part, nous étudions le checkpointing dans le cas de programmes parallèles MPI avec des communications point-à-point. Nous proposons des techniques pour appliquer le checkpointing à ces programmes. Nous fournissons des éléments de preuve de correction de nos techniques et nous les expérimentons sur des codes représentatifs. Ce travail a été effectué dans le cadre du projet européen ``AboutFlow'' / The adjoint mode of Algorithmic Differentiation (AD) is particularly attractive for computing gradients. However, this mode needs to use the intermediate values of the original simulation in reverse order at a cost that increases with the length of the simulation. AD research looks for strategies to reduce this cost, for instance by taking advantage of the structure of the given program. In this work, we consider on one hand the frequent case of Fixed-Point loops for which several authors have proposed adapted adjoint strategies. Among these strategies, we select the one introduced by B. Christianson. We specify further the selected method and we describe the way we implemented it inside the AD tool Tapenade. Experiments on a medium-size application shows a major reduction of the memory needed to store trajectories. On the other hand, we study checkpointing in the case of MPI parallel programs with point-to-point communications. We propose techniques to apply checkpointing to these programs. We provide proof of correctness of our techniques and we experiment them on representative CFD codes

Many different message-oriented middlewares (MOM) exist today on the market. With different underlying designs, purposes and features, it is not easy to make the right choice of MOM for your system. The IT company Nordicstation is in such a situation where they need to make a choice of MOM. They currently own a system called Sharelock which generates reports about violations of financial assets trad- ing. They wish to make the process of generating a report more asynchronous by using a MOM and they have a couple of requirements. This thesis was carried out with the purpose of finding the most suitable products for the system in mind, compare their features, performance, licenses and ease-of- use. This was accomplished by analysing their installation process, monitoring in- terfaces, documentation on their websites and performance in a simple throughput test. The results showed that RabbitMQ was the strongest candidate. It had good per- formance, an attractive web interface for monitoring, an easy installation and it also offered commercial support. Apache Artemis was also found to be an attractive choice but it did not have a web interface which made it hard to manage the system. / Idag finns det många meddelandebaserade mellanprogramvaror (MOM) på marknaden. Dessa har olika grundläggande designer, syften och kännetecken vilket kan göra det svårt att göra ett bra val av MOM för ett visst system. IT-företaget Nordicstation befinner sig i en sådan situation där de behöver göra ett val av MOM. De har ett system som kallas Sharelock som genererar rapporter om regelbrott in- om värdepappershandeln. De vill göra processen mer asynkron genom att använda en MOM och de har ett antal krav på produkten. Det här examensarbetet utfördes med syftet att hitta de mest passande produkter- na för det tänka systemet, jämföra deras särdrag, prestanda, licenser och anvä- ndarvänlighet. Detta gjordes genom att analysera deras installationsprocess, övervakningsgränssnitt, dokumentation på deras hemsida och prestanda i ett en- kelt prestandatest. Resultaten visade att RabbitMQ var den starkaste kandidaten. Den hade bra prestanda, ett attraktivt webbgränssnitt, en enkel installation och den erbjöd också kommersiell support åt kunder. Apache Artemis var också ett attraktivt val men den hade inget webbgränssnitt vilket gjorde det svårt att övervaka och hantera sys- temet.