Os compósitos cerâmicos de alto desempenho têm sido objeto de frequentes estudos nas últimas décadas, visando à melhora das propriedades mecânicas e ao aumento da sua gama de aplicações em produtos tecnológicos. Este trabalho consistiu em estudar a preparação, a sinterização convencional e não convencional e as propriedades mecânicas e tribológicas resultantes de dois sistemas compósitos: t-ZrO2/Al2O3 e t-ZrO2/Al2O3-NbC. No sistema t-ZrO2/Al2O3 foram estudadas as composições de 0, 5 e 15% em volume de Al2O3 usando pós comerciais. No sistema t-ZrO2/Al2O3-NbC, foi usado um pó nanocristalino de Al2O3-NbC, obtido por moagem reativa de alta energia e adicionado na proporção de 5% em volume à matriz de t-ZrO2. Os pós foram prensados uniaxial e isostaticamente e sinterizados em forno convencional e pelas técnicas de flash sintering (FS) (t-ZrO2/Al2O3) e spark plasma sintering (SPS) (t-ZrO2/Al2O3-NbC). Os compósitos t-ZrO2/Al2O3 sinterizados convencionalmente e t-ZrO2/Al2O3-NbC sinterizados convencionalmente e por SPS foram caracterizados por medidas de densidade aparente, dilatometria, microscopia eletrônica de varredura (MEV), e medidas de propriedades mecânicas: dureza, módulo de Young e tenacidade à fratura. Os compósitos t-ZrO2/Al2O3 sinterizados por FS foram caracterizados por medidas de densidade aparente, dilatometria in situ e MEV. Os nanocompósitos de t-ZrO2/Al2O3-NbC foram também caracterizados quanto à resistência ao desgaste pelo método esfera-no-disco, utilizando esferas de Al2O3 e WC-6%Co como contramateriais. Os resultados mostraram que a moagem reativa de alta energia foi completa e efetiva na obtenção de pós nanométricos de Al2O3-NbC, com tamanhos de cristalito de 9,1 nm para Al2O3 e 9,7 nm para o NbC. A desaglomeração posterior à moagem de alta energia foi eficaz na redução do tamanho de aglomerados. Os compósitos t-ZrO2/Al2O3 e t-ZrO2/Al2O3-NbC sinterizados convencionalmente e ZrO2/Al2O3-NbC sinterizados por SPS mostraram alta densificação (>97% DT e boas propriedades mecânicas. Os nanocompósitos de t-ZrO2/Al2O3 sinterizados por FS apresentaram uma densificação ultrarrápida (< 1 min) com retração linear superior às amostras sinterizadas em forno convencional, ocorrente a temperaturas inferiores a 1000°C, com densidades relativas superiores a 90% DT em algumas composições. Os nanocompósitos de t-ZrO2/Al2O3-NbC apresentaram propriedades competitivas entre os compósitos sinterizados convencionalmente e por SPS, com dureza e tenacidade à fratura superiores às da t-ZrO2 monolítica. A resistência ao desgaste desses nanocompósitos sinterizados convencionalmente, no entanto, foi notadamente superior à dos sinterizados por SPS. A oxidação do NbC nos compósitos sinterizados convencionalmente influiu negativamente nas propriedades, levando à sugestão de uma \"janela\" de temperaturas em que a sinterização do nanocompósito de t-ZrO2/Al2O3-NbC seja interessante sem a degradação das propriedades mecânicas. Os resultados permitiram concluir que os materiais estudados apresentam potencial para aplicações industriais que requerem cerâmicas de alto desempenho mecânico e de resistência ao desgaste. / High performance ceramic composites have been the subject of frequent studies in recent decades, aiming at improving mechanical properties and increasing their range of applications in technological products. This work consisted in studying the preparation, the conventional and non-conventional sintering and the mechanical properties resulting from two t-ZrO2 matrix composites: the t-ZrO2/Al2O3 system and the t-ZrO2/Al2O3-NbC system. In the t-ZrO2/Al2O3 system, the compositions of 0, 5 and 15% by volume of Al2O3 using commercial powders were studied, while in the t-ZrO2/Al2O3-NbC system, an Al2O3-NbC nanocrystalline powder obtained by high energy reactive milling, deagglomerated, leached in HCl and added in the proportion of 5% by volume to the t-ZrO2 matrix. The obtained powders were uniaxially and isostatically pressed and sintered in conventional furnace and using flash sintering (t-ZrO2/Al2O3) and spark plasma sintering (SPS) (t-ZrO2/Al2O3-NbC). Conventionally sintered t-ZrO2/Al2O3 and conventionally sintered t-ZrO2/Al2O3-NbC composites were characterized by measurements of apparent density, dilatometry, SEM, and mechanical properties: hardness, Young\'s modulus and fracture toughness. The t-ZrO2/Al2O3 composites sintered by FS were characterized by measurements of apparent density, in situ dilatometry and SEM. t-ZrO2/Al2O3-NbC nanocomposites were also characterized for wear strength by the ball-in-disc method, using Al2O3 and WC-6%Co beads as countermaterials. The results showed that the high energy reactive milling was complete and effective in obtaining nanometric powders of Al2O3-NbC, with crystallite sizes equal to 9.1 and 9.7 nm, for Al2O3 and NbC, respectively. The deagglomeration after high energy reactive milling was effective in reducing the size of agglomerates. Conventionally sintered t-ZrO2/Al2O3 and t-ZrO2/Al2O3-NbC composites and SPS-sintered t-ZrO2/Al2O3-NbC showed high densification (> 97% TD), good dispersion of the inclusions in the matrix and good mechanical properties. The t-ZrO2/Al2O3 nanocomposites sintered by FS presented an ultrafast densification (<1 min) with linear shrinkage superior to the sintered samples in conventional furnace, occurring at temperatures lower than 1000°C, with relative densities higher than 90% TD in some compositions. The t-ZrO2/Al2O3-NbC nanocomposites presented competitive properties between conventionally sintered and SPS-sintered composites with higher hardness and fracture toughness than monolithic t-ZrO2. The wear resistance of these conventionally sintered nanocomposites, however, was markedly higher than those of SPS-sintered ones. The oxidation of NbC in the composites sintered conventionally influenced negatively the properties, leading to the suggestion of a \"window\" of temperatures in which the sintering of the t-ZrO2/Al2O3-NbC nanocomposite is interesting without the degradation of the mechanical properties. The results allowed concluding that the studied materials present potential for industrial applications that require high mechanical performance and wear resistance ceramics.

Este trabalho compara o desempenho e a estabilidade de dois arcabouços para o processamento de Big Data: Apache Spark e High Performance Analytics Toolkit (HPAT). A comparação foi realizada usando duas aplicações: soma dos elementos de um vetor unidimensional e o algoritmo de clusterização K-means. Os experimentos foram realizados em ambiente distribuído e com memória compartilhada com diferentes quantidades e configurações de máquinas virtuais. Analisando os resultados foi possível concluir que o HPAT tem um melhor desempenho em relação ao Apache Spark nos nossos casos de estudo. Também realizamos uma análise dos dois arcabouços com a presença de falhas. / This work compares the performance and stability of two Big Data processing tools: Apache Spark and High Performance Analytics Toolkit (HPAT). The comparison was performed using two applications: a unidimensional vector sum and the K-means clustering algorithm. The experiments were performed in distributed and shared memory environments with different numbers and configurations of virtual machines. By analyzing the results we are able to conclude that HPAT has performance improvements in relation to Apache Spark in our case studies. We also provide an analysis of both frameworks in the presence of failures.

La production de nanoparticules (NPs) par des décharges spark en phase liquide permet d’atteindre des rendements jusqu’à présent inégalés de l’ordre de quelques centièmes de milligramme par joule. Même si l’essentiel de l’énergie est dissipé dans la formation de la décharge, l’érosion des électrodes métalliques permet la production efficace de NPs. La nature des NPs formées est largement tributaire du liquide diélectrique dans lequel la décharge est réalisée. Il est ainsi possible de contrôler les nanoparticules produites en choisissant de manière ad hoc les électrodes et le liquide. Nous cherchons dans ce travail à comprendre les mécanismes qui ont lieu durant la décharge en étudiant différents cas d’élaboration de NPs soit d’alliages soit de composites. Les NPs synthétisées auront des applications dans divers domaines selon le(s) matériau(x) choisi(s). À l’aide d’un générateur d’impulsions nanosecondes, les décharges ont été créées en appliquant une impulsion de haute tension (10 kV- 200ns- 10 Hz) entre deux électrodes immergées dans de l’azote liquide. Trois systèmes principaux ont été étudiés : Si-Sn, Cu-Zn et Cu-Ag. Les tests ont été réalisés avec des électrodes pures ou avec des alliages contenant les 2 éléments en proportions variables pour améliorer notre compréhension sur la manière dont les nanoparticules sont formées. Des analyses en microscopie électronique en transmission à haute résolution (HRTEM), en spectroscopie des rayons X à dispersion d'énergie (EDX), en spectroscopie de perte d'énergie des électrons (EELS) et des analyses de micro-diffraction ont été menées pour caractériser les NPs synthétisées (morphologie, cristallinité, composition chimique, etc.). Enfin, des mesures de spectroscopie d’émission optique résolues dans le temps ont été réalisées pour disposer d’informations sur l’évolution temporelle des raies émises au cours de la décharge et ainsi sur les conditions qui prévalent dans le plasma / Discharges in liquids offer a simple way to synthesize nanoparticles at high rate and low cost. When spark discharges are ignited in a dielectric liquid, a strong heating of the electrode material occurs, producing a metallic vapor from which nanoparticles grow by condensation. Even if most of the energy is dissipated in the formation of the discharge, the erosion of the metal electrodes allows the efficient production of NPs. The nature of the NPs formed is largely dependent on the dielectric liquid in which the discharge is performed. It is thus possible to control the nanoparticles produced by choosing the electrodes and the liquid in an appropriate manner. We seek in this work to understand the mechanisms that take place during the discharge by studying different cases of elaboration of NPs either alloys or composites. The synthesized NPs will have applications in various fields depending on the material(s) chosen. Using a nanosecond pulse generator, the discharges were created by applying a high voltage pulse (10 kV-200ns-10 Hz) between two electrodes immersed in liquid nitrogen. Three main systems have been studied: Si-Sn, Cu-Zn and Cu-Ag. The tests were performed with pure electrodes or alloys containing the 2 elements in varying proportions to improve our understanding of how nanoparticles are formed. High resolution transmission electron microscopy (HRTEM), energy dispersive X-ray spectroscopy (EDX), electron energy loss spectroscopy (EELS) and micro-diffraction analyzes were carried out to characterize the synthesized NPs (morphology, crystallinity, chemical composition, etc.). Finally, time-resolved optical emission spectroscopy measurements were performed to obtain information on the temporal evolution of the lines emitted during the discharge and thus on the conditions that prevail in the plasma

Le fonctionnement en régime très pauvre ou avec forts taux d'EGR des moteurs à allumage commandé (MAC) permet de réduire efficacement les émissions de CO2 et de Nox ; cependant ces stratégies se heurtent à l'augmentation des variabilités cycliques. Ces dernières sont principalement dues à la phase d'allumage qui devient critique de dilution. Le modèle ECFM-LES actuellement utilisé à IFPEn, basé sur la notion de densité de surface de flamme, est insuffisant pour décrire l'allumage dans ces conditions critiques. Dans ces travaux, l'approche TF-LES est adoptée, l'allumage étant alors décrit par un emballement cinétique des réactions chimiques lors d'une élévations locale de la température. Ces travaux définissent et évaluent une stratégie de simulation pour TF-LES en configuration moteur, qui permette une prédiction fine des allumages critiques et de la propagation turbulente de la flamme, afin de décrire le cycle moteur complet.Dans une première partie, des DNS d'allumages turbulents ont été réalisées, en modélisant la phase d'allumage par un dépôt d'énergie thermique (Lacaze et al., (2009)). Les calculs ont simulé les expériences d'allumage de Cardin et al. (2013), dans lesquelles l'énergie minimum d'allumage (MIE) d'un mélange mtéhane-air a été mesuré, pour différentes richesses pauvres et sous différentes intensités turbulentes. L'objectif principal des simulations a été de déterminer les paramètres numériques et physiques du modèle permettant de reproduire les allumages de l'expérience. Deux types de schémas cinétiques ont été évalués : un schéma simplifié et un schéma analytique (ARC), ce dernier reproduisant et les délais d'auto-allumage et la vitesse de flamme laminaire. Les résultats ont permis de définir des critères d'allumage et de mettre en évidence les différentes prédiction d'allumage avec les deux types de schémas cinétiques. Les résultats ont été également démontré que l'approche choisie permettait de prédire les bons niveaux d'énergie pour les allumages laminaires et à faible nombres de Kalovitz (Ka<10). Aux plus hauts nombres de Karlovitz, il a été montré que le modèle ED était insuffisant pour prédire les énergie d'allumage et qu'une description plus fine du dépôt d'énergie est nécessaire.Dans la seconde partie des travaux, un modèle de plissement dynamique (Wang et al., 2012) a été étudié, afin de décrire le développement hors-équilibre de la flamme dans la phase de propagation turbulente. Des études sur des flammes sphériques laminaires ont d'abord été menées. Ensuite, les premiers tests de configuration moteur ayant révélé des incompatibilités du modèle, des modifications ont été proposées. Le modèle de plissement dynamique modifié a été finalement évalué sur la configuration moteur ICAMDAC. Les résultats obtenus ont été comparés aux résultats obtenus par Robert et al. (2015) avec le modèle ECFM-LES, qui utilise une équation de transport de densité de surface de flamme décrivant le plissement hors-équilibre de la flamme. Les résultats obtenus avec le plissement dynamique sont en très bon accord avec ceux du modèles ECFM-LES, démontrant ainsi la capacité du modèle dynamique à prédire des valeurs de plissement hors-équilibre. D'autre part, le modèle dynamique s'ajustant automatiquement aux conditions de turbulence de l'écoulement, nul besoin n'est d'ajuster la constante de modélisation en fonction du régime moteur, comme c'est le cas pour l'équation de transport de la densité de surface de flamme. / The use of lean equivalence ratios or high EGR rates in spark ignition engines (SIE) enables to optimize CO2 and NOx emissions; however too important dilution rates leads to increased cycle-to-cycle variability. These latter are mostly due to the ignition phase, which becomes critical when dilution rates are important and requires high ignition energy. The ECFM-LES model currently used in IFPEN, which is based on the flame surface density concept, is not sufficient to describe ignition in these critical conditions. The TF-LES approach was chosen in this study, principally because it directly resolved chemistry and can thus model ignition via a local raise of the temperature. The present work defines and evaluates a simulation strategy for TF-LES in SIE configurations, that enables a fine prediction of critical ignitions and of the turbulent flame propagation.In the first part, DNS of turbulent ignition were performed. The ignition phase was modeled using a thermal energy deposit (ED model, Lacaze et al.). Simulations reproduced the ignition experiments of Cardin et al. who determined the minimum ignition energy (MIE) of lean premixed methane/air mixtures, for different turbulence characteristics. The main purpose of the study was to determine the numerical and physical model parameters, which enable to reproduce Cardin et al. experiments. Two types of kinetic schemes were evaluated: a simplified kinetic scheme and an analytical kinetic scheme (ARC), that can predict both the auto-ignition delays and the laminar flame speed, while keeping affordable CPU times. Results analysis enabled to define ignition criteria and to highlight the differences in terms of ignition prediction using the two kinetic schemes. Results also demonstrated that the chosen approach could recover correct levels of ignition energy for laminar and low Karlovitz number cases (Ka<10). For higher Karlovitz number cases, the ED model was found to be insufficient to predict the ignition and a finer description of the energy deposit is required.In the second part, a dynamic wrinkling model (Wang et al., 2012) was studied to describe the out-of-equilibrium behavior of the flame during the propagation phase. Studies on laminar spherical flames were first performed, to assess the laminar degeneration of the model. Then, as first tests in an engine configuration have revealed incompatibilities of the model, modifications were proposed. The modified dynamic model was finally tested in the ICAMDAC engine configuration. Results of the simulations were compared against previous results of Robert et al. obtained with the ECFM-LES model using a transport equation for the flame surface density that can describe the out-of-equilibrium wrinkling of the flame. Results obtained with the dynamic model are in very good agreement with the ones of Robert et al., thus demonstrating the ability of the dynamic model to predict out-of-equilibrium values in the engine configuration. Besides, the dynamic model self-adapts to the turbulence conditions, hence does not require any model parameter adjustment, as is it the case for models based on the flame surface density transport equation.

This thesis addresses the challenges of large software and data-intensive systems. We will discuss a Big Data software that consists of quite a bit of Linux configuration, some Scala coding and a set of frameworks that work together to achieve the smooth performance of the system. Moreover, the thesis focuses on the Apache Spark framework and the challenging of measuring the lazy evaluation of the transformation operations of Spark. Investigating the challenges are essential for the performance engineers to increase their ability to study how the system behaves and take decisions in early design iteration. Thus, we made some experiments and measurements to achieve this goal. In addition to that, and after analyzing the result we could create a formula that will be useful for the engineers to predict the performance of the system in production.

Nowadays, the amount of data generated by users within an Internet product is increasing exponentially, for instance, clickstream for a website application from millions of users, geospatial information from GIS-based APPs of Android and IPhone, or sensor data from cars or any electronic equipment, etc. All these data may be yielded billions every day, which is not surprisingly essential that insights could be extracted or built. For instance, monitoring system, fraud detection, user behavior analysis and feature verification, etc.Nevertheless, technical issues emerge accordingly. Heterogeneity, massiveness and miscellaneous requirements for taking use of the data from different dimensions make it much harder when it comes to the design of data pipelines, transforming and persistence in data warehouse. Undeniably, there are traditional ways to build ETLs from mainframe [1], RDBMS, to MapReduce and Hive. Yet with the emergence and popularization of Spark framework and AWS, this procedure could be evolved to a more robust, efficient, less costly and easy-to-implement architecture for collecting, building dimensional models and proceeding analytics on massive data. With the advantage of being in a car transportation company, billions of user behavior events come in every day, this paper contributes to an exploratory way of building and optimizing ETL pipelines based on AWS and Spark, and compare it with current main Data pipelines from different aspects. / Mängden data som genereras internet-produkt-användare ökar lavinartat och exponentiellt. Det finns otaliga exempel på detta; klick-strömmen från hemsidor med miljontals användare, geospatial information från GISbaserade Android och iPhone appar, eller från sensorer på autonoma bilar.Mängden händelser från de här typerna av data kan enkelt uppnå miljardantal dagligen, därför är det föga förvånande att det är möjligt att extrahera insikter från de här data-strömmarna. Till exempel kan man sätta upp automatiserade övervakningssystem eller kalibrera bedrägerimodeller effektivt. Att handskas med data i de här storleksordningarna är dock inte helt problemfritt, det finns flertalet tekniska bekymmer som enkelt kan uppstå. Datan är inte alltid på samma form, den kan vara av olika dimensioner vilket gör det betydligt svårare att designa en effektiv data-pipeline, transformera datan och lagra den persistent i ett data-warehouse. Onekligen finns det traditionella sätt att bygga ETL’s på från mainframe [1], RDBMS, till MapReduce och Hive. Dock har det med upptäckten och ökade populariteten av Spark och AWS blivit mer robust, effektivt, billigare och enklare att implementera system för att samla data, bygga dimensions-enliga modeller och genomföra analys av massiva data-set. Den här uppsatsen bidrar till en ökad förståelse kring hur man bygger och optimerar ETL-pipelines baserade på AWS och Spark och jämför med huvudsakliga nuvarande Data-pipelines med hänsyn till diverse aspekter. Uppsatsen drar nytta av att ha tillgång till ett massivt data-set med miljarder användar-events genererade dagligen från ett bil-transport-bolag i mellanöstern.

L’intérêt de ces travaux est de proposer un nouveau composite BaAl2Si2O8 (BAS) renforcé par des fibres d’alumine ayant des propriétés physiques similaires tout en étant plus réfractaire que les composites SiO2 / SiO2. La forme cristalline hexagonale du BAS est la forme stable à haute température. Cependant, elle est métastable en dessous de 1590 °C et il est donc nécessaire de la stabiliser pour éviter les transformations cristallines. La stabilisation de la forme hexagonale par substitution atomique, notamment par du rubidium à hauteur de 5 % atomique a été la solution retenue. Concernant le composite à matrice BAS, le choix de fibres d’alumine est motivée par la compatibilité physico-chimique BAS / alumine. Diverses voies d’élaboration de la matrice BAS et du composite BAS / alumine ont été explorées. La voie d’élaboration par « reactive spark plasma sintering » (R-SPS) apporte un gain important en termes de réduction du temps d’élaboration et de rendement. Ce gain de temps évite donc l’exposition du BAS aux hautes températures et le risque de transformation de la phase hexagonale en monoclinique. Cependant la mise en forme par SPS de matériaux oxydes, dont le BAS, est confrontée à l’existence de gradients thermiques importants au sein de l’échantillon. C’est pourquoi, la mise en oeuvre d’un moule chauffant est développée comme une alternative au SPS. Ce système, en cours d’évaluation, devrait permettre l’utilisation de cycles thermiques équivalents à ceux du SPS, tout en limitant fortement les gradients thermiques. / The main purpose of the present work is to propose a new BaAl2Si2O8 (BAS) composite reinforced with alumina fibers exhibiting similar physical properties but a higher refractoriness than SiO2 / SiO2 composites. The hexagonal crystal form of BAS is the stable one at high temperatures. However, it is metastable below 1590 °C and it is therefore necessary to stabilize it in order to prevent crystalline transformations. The stabilization of the hexagonal form by atomic substitution, including rubidium at 5 atomic % has been chosen. Regarding the matrix BAS composite, the alumina fibers selection has been justified by their low physical and chemical reactivity with this material. Several elaboration methods of the BAS matrix and of the BAS / alumina composite have been investigated. The development by "reactive spark plasma sintering" (R-SPS) brings an important benefit in terms of reduced elaboration time and yield. This saving time thus limits the BAS exposure to high temperatures and the risk of transformation into monoclinic. Nevertheless, the SPS shaping of oxide materials, including the BAS, is confronted with the presence of important thermal gradient within the sample. This is why shaping in a heating mold is currently in progress, as an alternative to the SPS. This system should allow the use the same thermal cycles as for SPS, but with lower thermal gradient.

Le "Spark Plasma Sintering" est un procédé innovant qui permet de densifier, assembler, forger... tous types de matériaux avec des cinétiques très rapides. Cependant, il nécessite des améliorations pour le contrôle des températures, l'homogénéité microstructurale pour des pièces de formes complexes... et de productivité industrielle. Pour résoudre ces problèmes, un modèle électro-thermo-mécanique-microstructural est identifié : i) pour la partie Thermo-Electrique une instrumentation fine a permis par une approche inverse d'évaluer les résistances de contacts, ii) par essais in-situ de fluage et de compression pour la partie mécanique-microstructurale. Il a permis de trouver des solutions pertinentes pour élaborer des pièces - de microstructure homogène - simultanément en grand nombre (modification du passage du courant électrique) - de formes complexes (intervention de pièces sacrificielles). / The "Spark Plasma Sintering" process allows very high consolidation kinetics (densification, assembly, forging) of materials (powder, porous, nanostructured). However, some difficulties remains on this innovative process, particularly in terms of temperature control, microstructural homogeneity especially for complex shapes ... and industrial productivity. To solve these problems, an electro-thermo-mechanical-microstructural model is identified: i) using a thin instrumentation of the machine for the Thermo-Electric part including a reverse approach to evaluate the contact resistances, ii) by in situ creep and compression tests for mechanical microstructural-part. The resulting model has helped to find solutions for microstructural homogenization of the parts, for simultaneously densify of large numbers of parts (modifying the flow of electric current) and/or complex shapes (intervention sacrificial parts).

Os compósitos cerâmicos de alto desempenho têm sido objeto de frequentes estudos nas últimas décadas, visando à melhora das propriedades mecânicas e ao aumento da sua gama de aplicações em produtos tecnológicos. Este trabalho consistiu em estudar a preparação, a sinterização convencional e não convencional e as propriedades mecânicas e tribológicas resultantes de dois sistemas compósitos: t-ZrO2/Al2O3 e t-ZrO2/Al2O3-NbC. No sistema t-ZrO2/Al2O3 foram estudadas as composições de 0, 5 e 15% em volume de Al2O3 usando pós comerciais. No sistema t-ZrO2/Al2O3-NbC, foi usado um pó nanocristalino de Al2O3-NbC, obtido por moagem reativa de alta energia e adicionado na proporção de 5% em volume à matriz de t-ZrO2. Os pós foram prensados uniaxial e isostaticamente e sinterizados em forno convencional e pelas técnicas de flash sintering (FS) (t-ZrO2/Al2O3) e spark plasma sintering (SPS) (t-ZrO2/Al2O3-NbC). Os compósitos t-ZrO2/Al2O3 sinterizados convencionalmente e t-ZrO2/Al2O3-NbC sinterizados convencionalmente e por SPS foram caracterizados por medidas de densidade aparente, dilatometria, microscopia eletrônica de varredura (MEV), e medidas de propriedades mecânicas: dureza, módulo de Young e tenacidade à fratura. Os compósitos t-ZrO2/Al2O3 sinterizados por FS foram caracterizados por medidas de densidade aparente, dilatometria in situ e MEV. Os nanocompósitos de t-ZrO2/Al2O3-NbC foram também caracterizados quanto à resistência ao desgaste pelo método esfera-no-disco, utilizando esferas de Al2O3 e WC-6%Co como contramateriais. Os resultados mostraram que a moagem reativa de alta energia foi completa e efetiva na obtenção de pós nanométricos de Al2O3-NbC, com tamanhos de cristalito de 9,1 nm para Al2O3 e 9,7 nm para o NbC. A desaglomeração posterior à moagem de alta energia foi eficaz na redução do tamanho de aglomerados. Os compósitos t-ZrO2/Al2O3 e t-ZrO2/Al2O3-NbC sinterizados convencionalmente e ZrO2/Al2O3-NbC sinterizados por SPS mostraram alta densificação (&gt;97% DT e boas propriedades mecânicas. Os nanocompósitos de t-ZrO2/Al2O3 sinterizados por FS apresentaram uma densificação ultrarrápida (&lt; 1 min) com retração linear superior às amostras sinterizadas em forno convencional, ocorrente a temperaturas inferiores a 1000°C, com densidades relativas superiores a 90% DT em algumas composições. Os nanocompósitos de t-ZrO2/Al2O3-NbC apresentaram propriedades competitivas entre os compósitos sinterizados convencionalmente e por SPS, com dureza e tenacidade à fratura superiores às da t-ZrO2 monolítica. A resistência ao desgaste desses nanocompósitos sinterizados convencionalmente, no entanto, foi notadamente superior à dos sinterizados por SPS. A oxidação do NbC nos compósitos sinterizados convencionalmente influiu negativamente nas propriedades, levando à sugestão de uma \"janela\" de temperaturas em que a sinterização do nanocompósito de t-ZrO2/Al2O3-NbC seja interessante sem a degradação das propriedades mecânicas. Os resultados permitiram concluir que os materiais estudados apresentam potencial para aplicações industriais que requerem cerâmicas de alto desempenho mecânico e de resistência ao desgaste. / High performance ceramic composites have been the subject of frequent studies in recent decades, aiming at improving mechanical properties and increasing their range of applications in technological products. This work consisted in studying the preparation, the conventional and non-conventional sintering and the mechanical properties resulting from two t-ZrO2 matrix composites: the t-ZrO2/Al2O3 system and the t-ZrO2/Al2O3-NbC system. In the t-ZrO2/Al2O3 system, the compositions of 0, 5 and 15% by volume of Al2O3 using commercial powders were studied, while in the t-ZrO2/Al2O3-NbC system, an Al2O3-NbC nanocrystalline powder obtained by high energy reactive milling, deagglomerated, leached in HCl and added in the proportion of 5% by volume to the t-ZrO2 matrix. The obtained powders were uniaxially and isostatically pressed and sintered in conventional furnace and using flash sintering (t-ZrO2/Al2O3) and spark plasma sintering (SPS) (t-ZrO2/Al2O3-NbC). Conventionally sintered t-ZrO2/Al2O3 and conventionally sintered t-ZrO2/Al2O3-NbC composites were characterized by measurements of apparent density, dilatometry, SEM, and mechanical properties: hardness, Young\'s modulus and fracture toughness. The t-ZrO2/Al2O3 composites sintered by FS were characterized by measurements of apparent density, in situ dilatometry and SEM. t-ZrO2/Al2O3-NbC nanocomposites were also characterized for wear strength by the ball-in-disc method, using Al2O3 and WC-6%Co beads as countermaterials. The results showed that the high energy reactive milling was complete and effective in obtaining nanometric powders of Al2O3-NbC, with crystallite sizes equal to 9.1 and 9.7 nm, for Al2O3 and NbC, respectively. The deagglomeration after high energy reactive milling was effective in reducing the size of agglomerates. Conventionally sintered t-ZrO2/Al2O3 and t-ZrO2/Al2O3-NbC composites and SPS-sintered t-ZrO2/Al2O3-NbC showed high densification (&gt; 97% TD), good dispersion of the inclusions in the matrix and good mechanical properties. The t-ZrO2/Al2O3 nanocomposites sintered by FS presented an ultrafast densification (&lt;1 min) with linear shrinkage superior to the sintered samples in conventional furnace, occurring at temperatures lower than 1000°C, with relative densities higher than 90% TD in some compositions. The t-ZrO2/Al2O3-NbC nanocomposites presented competitive properties between conventionally sintered and SPS-sintered composites with higher hardness and fracture toughness than monolithic t-ZrO2. The wear resistance of these conventionally sintered nanocomposites, however, was markedly higher than those of SPS-sintered ones. The oxidation of NbC in the composites sintered conventionally influenced negatively the properties, leading to the suggestion of a \"window\" of temperatures in which the sintering of the t-ZrO2/Al2O3-NbC nanocomposite is interesting without the degradation of the mechanical properties. The results allowed concluding that the studied materials present potential for industrial applications that require high mechanical performance and wear resistance ceramics.

Este trabalho compara o desempenho e a estabilidade de dois arcabouços para o processamento de Big Data: Apache Spark e High Performance Analytics Toolkit (HPAT). A comparação foi realizada usando duas aplicações: soma dos elementos de um vetor unidimensional e o algoritmo de clusterização K-means. Os experimentos foram realizados em ambiente distribuído e com memória compartilhada com diferentes quantidades e configurações de máquinas virtuais. Analisando os resultados foi possível concluir que o HPAT tem um melhor desempenho em relação ao Apache Spark nos nossos casos de estudo. Também realizamos uma análise dos dois arcabouços com a presença de falhas. / This work compares the performance and stability of two Big Data processing tools: Apache Spark and High Performance Analytics Toolkit (HPAT). The comparison was performed using two applications: a unidimensional vector sum and the K-means clustering algorithm. The experiments were performed in distributed and shared memory environments with different numbers and configurations of virtual machines. By analyzing the results we are able to conclude that HPAT has performance improvements in relation to Apache Spark in our case studies. We also provide an analysis of both frameworks in the presence of failures.