Conselho Nacional de Desenvolvimento Científico e Tecnológico / A obtenção de imagens usando tomografia computadorizada revolucionou o diagnóstico de doenças na medicina e é usada amplamente em diferentes áreas da pesquisa científica. Como parte do processo de obtenção das imagens tomográficas tridimensionais um conjunto de radiografias são processadas por um algoritmo computacional, o mais usado atualmente é o algoritmo de Feldkamp, David e Kress (FDK). Os usos do processamento paralelo para acelerar os cálculos em algoritmos computacionais usando as diferentes tecnologias disponíveis no mercado têm mostrado sua utilidade para diminuir os tempos de processamento. No presente trabalho é apresentada a paralelização do algoritmo de reconstrução de imagens tridimensionais FDK usando unidades gráficas de processamento (GPU) e a linguagem CUDA-C. São apresentadas as GPUs como uma opção viável para executar computação paralela e abordados os conceitos introdutórios associados à tomografia computadorizada, GPUs, CUDA-C e processamento paralelo. A versão paralela do algoritmo FDK executada na GPU é comparada com uma versão serial do mesmo, mostrando maior velocidade de processamento. Os testes de desempenho foram feitos em duas GPUs de diferentes capacidades: a placa NVIDIA GeForce 9400GT (16 núcleos) e a placa NVIDIA Quadro 2000 (192 núcleos). / The imaging using computed tomography has revolutionized the diagnosis of diseases in medicine and is widely used in different areas of scientific research. As part of the process to obtained three-dimensional tomographic images a set of x-rays are processed by a computer algorithm, the most widely used algorithm is Feldkamp, David and Kress (FDK). The use of parallel processing to speed up calculations on computer algorithms with the different available technologies, showing their usefulness to decrease processing times. In the present paper presents the parallelization of the algorithm for three-dimensional image reconstruction FDK using graphics processing units (GPU) and CUDA-C. GPUs are shown as a viable option to perform parallel computing and addressed the introductory concepts associated with computed tomographic, GPUs, CUDA-C and parallel processing. The parallel version of the FDK algorithm is executed on the GPU and compared to a serial version of the same, showing higher processing speed. Performance tests were made in two GPUs with different capacities, the NVIDIA GeForce 9400GT (16 cores) and NVIDIA GeForce 2000 (192 cores).

Conselho Nacional de Desenvolvimento Científico e Tecnológico / A obtenção de imagens usando tomografia computadorizada revolucionou o diagnóstico de doenças na medicina e é usada amplamente em diferentes áreas da pesquisa científica. Como parte do processo de obtenção das imagens tomográficas tridimensionais um conjunto de radiografias são processadas por um algoritmo computacional, o mais usado atualmente é o algoritmo de Feldkamp, David e Kress (FDK). Os usos do processamento paralelo para acelerar os cálculos em algoritmos computacionais usando as diferentes tecnologias disponíveis no mercado têm mostrado sua utilidade para diminuir os tempos de processamento. No presente trabalho é apresentada a paralelização do algoritmo de reconstrução de imagens tridimensionais FDK usando unidades gráficas de processamento (GPU) e a linguagem CUDA-C. São apresentadas as GPUs como uma opção viável para executar computação paralela e abordados os conceitos introdutórios associados à tomografia computadorizada, GPUs, CUDA-C e processamento paralelo. A versão paralela do algoritmo FDK executada na GPU é comparada com uma versão serial do mesmo, mostrando maior velocidade de processamento. Os testes de desempenho foram feitos em duas GPUs de diferentes capacidades: a placa NVIDIA GeForce 9400GT (16 núcleos) e a placa NVIDIA Quadro 2000 (192 núcleos). / The imaging using computed tomography has revolutionized the diagnosis of diseases in medicine and is widely used in different areas of scientific research. As part of the process to obtained three-dimensional tomographic images a set of x-rays are processed by a computer algorithm, the most widely used algorithm is Feldkamp, David and Kress (FDK). The use of parallel processing to speed up calculations on computer algorithms with the different available technologies, showing their usefulness to decrease processing times. In the present paper presents the parallelization of the algorithm for three-dimensional image reconstruction FDK using graphics processing units (GPU) and CUDA-C. GPUs are shown as a viable option to perform parallel computing and addressed the introductory concepts associated with computed tomographic, GPUs, CUDA-C and parallel processing. The parallel version of the FDK algorithm is executed on the GPU and compared to a serial version of the same, showing higher processing speed. Performance tests were made in two GPUs with different capacities, the NVIDIA GeForce 9400GT (16 cores) and NVIDIA GeForce 2000 (192 cores).

Avec l'émergence des bâtiments à haute efficacité énergétique, il est devenu indispensable de pouvoir prédire de manière fiable le comportement énergétique des bâtiments. Or, à l'heure actuelle, la prise en compte des effets thermo-aérauliques dans les modèles se cantonne le plus souvent à l'utilisation d'approches simplifiées voire empiriques qui ne sauraient atteindre la précision requise. Le recours à la simulation numérique des écoulements semble donc incontournable, mais il est limité par un coût calculatoire généralement prohibitif. L'utilisation conjointe d'approches innovantes telle que la méthode de Boltzmann sur gaz réseau (LBM) et d'outils de calcul massivement parallèles comme les processeurs graphiques (GPU) pourrait permettre de s'affranchir de ces limites. Le présent travail de recherche s'attache à en explorer les potentialités. La méthode de Boltzmann sur gaz réseau, qui repose sur une forme discrétisée de l'équation de Boltzmann, est une approche explicite qui jouit de nombreuses qualités : précision, stabilité, prise en compte de géométries complexes, etc. Elle constitue donc une alternative intéressante à la résolution directe des équations de Navier-Stokes par une méthode numérique classique. De par ses caractéristiques algorithmiques, elle se révèle bien adaptée au calcul parallèle. L'utilisation de processeurs graphiques pour mener des calculs généralistes est de plus en plus répandue dans le domaine du calcul intensif. Ces processeurs à l'architecture massivement parallèle offrent des performances inégalées à ce jour pour un coût relativement modéré. Néanmoins, nombre de contraintes matérielles en rendent la programmation complexe et les gains en termes de performances dépendent fortement de la nature de l'algorithme considéré. Dans le cas de la LBM, les implantations GPU affichent couramment des performances supérieures de deux ordres de grandeur à celle d'une implantation CPU séquentielle faiblement optimisée. Le mémoire de thèse présenté est constitué d'un ensemble de neuf articles de revues internationales et d'actes de conférences internationales (le dernier étant en cours d'évaluation). Dans ces travaux sont abordés les problématiques liées tant à l'implantation mono-GPU de la LBM et à l'optimisation des accès en mémoire, qu'aux implantations multi-GPU et à la modélisation des communications inter-GPU et inter-nœuds. En complément, sont détaillées diverses extensions à la LBM indispensables pour envisager une utilisation en thermo-aéraulique des bâtiments. Les cas d'études utilisés pour la validation des codes permettent de juger du fort potentiel de cette approche en pratique. / With the advent of low-energy buildings, the need for accurate building performance simulations has significantly increased. However, for the time being, the thermo-aeraulic effects are often taken into account through simplified or even empirical models, which fail to provide the expected accuracy. Resorting to computational fluid dynamics seems therefore unavoidable, but the required computational effort is in general prohibitive. The joint use of innovative approaches such as the lattice Boltzmann method (LBM) and massively parallel computing devices such as graphics processing units (GPUs) could help to overcome these limits. The present research work is devoted to explore the potential of such a strategy. The lattice Boltzmann method, which is based on a discretised version of the Boltzmann equation, is an explicit approach offering numerous attractive features: accuracy, stability, ability to handle complex geometries, etc. It is therefore an interesting alternative to the direct solving of the Navier-Stokes equations using classic numerical analysis. From an algorithmic standpoint, the LBM is well-suited for parallel implementations. The use of graphics processors to perform general purpose computations is increasingly widespread in high performance computing. These massively parallel circuits provide up to now unrivalled performance at a rather moderate cost. Yet, due to numerous hardware induced constraints, GPU programming is quite complex and the possible benefits in performance depend strongly on the algorithmic nature of the targeted application. For LBM, GPU implementations currently provide performance two orders of magnitude higher than a weakly optimised sequential CPU implementation. The present thesis consists of a collection of nine articles published in international journals and proceedings of international conferences (the last one being under review). These contributions address the issues related to single-GPU implementations of the LBM and the optimisation of memory accesses, as well as multi-GPU implementations and the modelling of inter-GPU and internode communication. In addition, we outline several extensions to the LBM, which appear essential to perform actual building thermo-aeraulic simulations. The test cases we used to validate our codes account for the strong potential of GPU LBM solvers in practice.

本研究要探討的是在新聞、論壇和社群媒體討論的相關訊息是否真的會影響匯率的運動的假設。對於這樣的研究目標，我們建立了一個實驗，首先以文字探勘技術應用在新聞、論壇與社群媒體來產生與匯率相關的數值表示。接著，機器學習技術應用於學習得到的數值表示和匯率波動之間的關係。最後，我們證明透過檢驗所獲得的關係的有效性的假設。在此研究中，我們提出一種兩階段的神經網路來學習與預測每日美金兌台幣匯率的走勢。不同於其他專注於新聞或者社群媒體的研究，我們將他們進行整合，並將論壇的討論納為輸入資料。不同的資料組合產生出多種觀點，而三個資料來源的不同組合可能會以不同的方式影響預測準確率。透過該方法，初步實驗的結果顯示此方法優於隨機漫步模型。 / This study wants to explore the hypothesis that the relevant information in the news, the posts in forums and discussions on the social media can really affect the daily movement of exchange rates. For such study objective, we set up an experiment, where the text mining technique is first applied to the news, the forum and the social media to generate numerical representations regarding the textual information relevant with the exchange rate. Then the machine learning technique is applied to learn the relationship between the derived numerical representations and the movement of exchange rates. At the end, we justify the hypothesis through examining the effectiveness of the obtained relationship. In this paper, we propose a hybrid neural networks to learn and forecast the daily movements of USD/TWD exchange rates. Different from other studies, which focus on news or social media, we integrate them and add the discussion of forum as input data. Different data combinations yield many views while different combination of three data sources might affect the forecasting accuracy rate in different ways. As a result of this method, the experiment result was better than random walk model.

The presented doctoral thesis aims at development a new efficient tool for optimization of uniformity of point samples. One of use-cases of these point sets is the usage as optimized sets of integration points in statistical analyses of computer models using Monte Carlo type integration. It is well known that the pursuit of uniformly distributed sets of integration points is the only possible way of decreasing the error of estimation of an integral over an unknown function. The tasks of the work concern a survey of currently used criteria for evaluation and/or optimization of uniformity of point sets. A critical evaluation of their properties is presented, leading to suggestions towards improvements in spatial and statistical uniformity of resulting samples. A refined variant of the general formulation of the phi optimization criterion has been derived by incorporating the periodically repeated design domain along with a scale-independent behavior of the criterion. Based on a notion of a physical analogy between a set of sampling points and a dynamical system of mutually repelling particles, a hyper-dimensional N-body system has been selected to be the driver of the developed optimization tool. Because the simulation of such a dynamical system is known to be a computationally intensive task, an efficient solution using the massively parallel GPGPU platform Nvidia CUDA has been developed. An intensive study of properties of this complex architecture turned out as necessary to fully exploit the possible solution speedup.

Continuous casting is a dominant production technology of steelmaking which is currently used for more that 95% of the world steel production. Mathematical modelling and optimal control of casting machine are crucial tasks in continuous steel casting which directly influence productivity and quality of produced steel, competitiveness of steelworks, safety of casting machine operation and its impact on the environment. This thesis concerns with the development and implementation of the numerical model of temperature field for continuously cast steel billets and its use for optimal control of the casting machine. The numerical model was developed and implemented in MATLAB. Due to computational demands the model was parallelized by means of the computation on graphics processing units NVIDIA with the computational architecture CUDA. Validation and verification of the model were performed with the use of operational data from Trinecke zelezarny steelworks. The model was then utilized as a part of the developed model-based predictive control system for the optimal control of dynamic situations in the casting machine operation. The behaviour of the developed control system was examined by means of dynamic model situations that have confirmed the ability of the implemented system to optimally control dynamic operations of the continuous casting machine. Both the numerical model of the temperature field and the model-based predictive control system have been implemented so that they can be modified for any casting machine and this allows for their prospective commercial applications.

Tato práce se zaměřuje na datové struktury pro reprezentaci řídkých blokových matic a s nimi spojených výpočetních algoritmů, jež jsem navrhl. Řídké blokové matice se vyskytují při řešení mnoha dílčích problémů jako například při řešení metody nejmenších čtverců. Nelineární metoda nejmenších čtverců (NLS) je často aplikována v robotice pro řešení problému lokalizace robota (SLAM) nebo v příbuzných úlohách 3D rekonstrukce v počítačovém vidění (BA), (SfM). Problémy konečných elementů (FEM) a parciálních diferenciálních rovnic (PDE) v oboru fyzikálních simulací můžou také mít blokovou strukturu. Většina existujících implementací řídké lineární algebry používají řídké matice s granularitou jednotlivých elementů a jen několik málo podporuje řídké blokové matice. To může být způsobeno složitostí blokových formátů, jež snižuje rychlost výpočtů, pokud bloky nejsou dost velké. Některé ze specializovaných NLS optimalizátorů v robotice a počítačovém vidění používají blokové matice jako interní reprezentaci, aby snížily cenu sestavování řídkých matic, ale nakonec tuto reprezentaci převedou na elementovou řídkou matici pro implementaci k řešení systémů rovnic. Existující implementace pro řídké blokové matice se většinou soustředí na jedinou operaci, často násobení matice vektorem. Řešení navržené v této disertaci pokrývá širší spektrum funkcí: implementovány jsou funkce pro efektivní sestavení řídké blokové matice, násobení matice vektorem nebo jinou maticí a nechybí ani řešení trojúhelníkových systémů nebo Choleského faktorizace. Tyto funkce mohou být snadno použity ke řešení systémů lineárních rovnic pomocí analytických nebo iterativních metod nebo k výpočtu vlastních čísel. Jsou zde popsány rychlé algoritmy pro hlavní procesor (CPU) i pro grafické akcelerátory (GPU). Navrhované algoritmy jsou integrovány v knihovně SLAM++ , jež řeší problém nelineárních nejmenších čtverců se zaměřením na problémy v robotice a počítačovém vidění. Je provedeno vyhodnocení na standardních datasetech kde navrhované metody dosahují výrazně lepších výsledků než dosavadní metody popsané v literatuře -- a to bez kompromisů v přesnosti či obecnosti řešení.

La simulation est de plus en plus utilisée dans le domaine industriel du Contrôle Non Destructif. Elle est employée tout au long du processus de contrôle, que ce soit pour en accélérer la mise au point ou en comprendre les résultats. Les travaux menés au cours de cette thèse présentent une méthode de calcul rapide de champ ultrasonore rayonné par un capteur multi-éléments dans une pièce isotrope, permettant un usage interactif des simulations. Afin de tirer parti des architectures parallèles communément disponibles, un modèle régulier (qui limite au maximum les branchements divergents) dérivé du modèle générique présent dans la plateforme logicielle CIVA a été mis au point. Une première implémentation de référence a permis de le valider par rapport aux résultats CIVA et d'analyser son comportement en termes de performances. Le code a ensuite été porté et optimisé sur trois classes d'architectures parallèles aujourd'hui disponibles dans les stations de calcul : le processeur généraliste central (GPP), le coprocesseur manycore (Intel MIC) et la carte graphique (nVidia GPU). Concernant le processeur généraliste et le coprocesseur manycore, l'algorithme a été réorganisé et le code implémenté afin de tirer parti des deux niveaux de parallélisme disponibles, le multithreading et les instructions vectorielles. Sur la carte graphique, les différentes étapes de simulation de champ ont été découpées en une série de noyaux CUDA. Enfin, des bibliothèques de calculs spécifiques à ces architectures, Intel MKL et nVidia cuFFT, ont été utilisées pour effectuer les opérations de Transformées de Fourier Rapides. Les performances et la bonne adéquation des codes produits ont été analysées en détail pour chaque architecture. Dans plusieurs cas, sur des configurations de contrôle réalistes, des performances autorisant l'interactivité ont été atteintes. Des perspectives pour traiter des configurations plus complexes sont dressées. Enfin la problématique de l'industrialisation de ce type de code dans la plateforme logicielle CIVA est étudiée. / The Non Destructive Testing field increasingly uses simulation.It is used at every step of the whole control process of an industrial part, from speeding up control development to helping experts understand results. During this thesis, a simulation tool dedicated to the fast computation of an ultrasonic field radiated by a phase array probe in an isotropic specimen has been developped. Its performance enables an interactive usage. To benefit from the commonly available parallel architectures, a regular model (aimed at removing divergent branching) derived from the generic CIVA model has been developped. First, a reference implementation was developped to validate this model against CIVA results, and to analyze its performance behaviour before optimization. The resulting code has been optimized for three kinds of parallel architectures commonly available in workstations: general purpose processors (GPP), manycore coprocessors (Intel MIC) and graphics processing units (nVidia GPU). On the GPP and the MIC, the algorithm was reorganized and implemented to benefit from both parallelism levels, multhreading and vector instructions. On the GPU, the multiple steps of field computing have been divided in multiple successive CUDA kernels.Moreover, libraries dedicated to each architecture were used to speedup Fast Fourier Transforms, Intel MKL on GPP and MIC and nVidia cuFFT on GPU. Performance and hardware adequation of the produced algorithms were thoroughly studied for each architecture. On multiple realistic control configurations, interactive performance was reached. Perspectives to adress more complex configurations were drawn. Finally, the integration and the industrialization of this code in the commercial NDT plateform CIVA is discussed.

The race to computing power increases every day in the simulation community. A few years ago, scientists have started to harness the computing power of Graphics Processing Units (GPUs) to parallelize their simulations. As with any parallel architecture, not only the simulation model implementation has to be ported to the new parallel platform, but all the tools must be reimplemented as well. In the particular case of stochastic simulations, one of the major element of the implementation is the pseudorandom numbers source. Employing pseudorandom numbers in parallel applications is not a straightforward task, and it has to be done with caution in order not to introduce biases in the results of the simulation. This problematic has been studied since parallel architectures are available and is called pseudorandom stream distribution. While the literature is full of solutions to handle pseudorandom stream distribution on CPU-based parallel platforms, the young GPU programming community cannot display the same experience yet. In this thesis, we study how to correctly distribute pseudorandom streams on GPU. From the existing solutions, we identified a need for good software engineering solutions, coupled to sound theoretical choices in the implementation. We propose a set of guidelines to follow when a PRNG has to be ported to GPU, and put these advice into practice in a software library called ShoveRand. This library is used in a stochastic Polymer Folding model that we have implemented in C++/CUDA. Pseudorandom streams distribution on manycore architectures is also one of our concerns. It resulted in a contribution named TaskLocalRandom, which targets parallel Java applications using pseudorandom numbers and task frameworks. Eventually, we share a reflection on the methods to choose the right parallel platform for a given application. In this way, we propose to automatically build prototypes of the parallel application running on a wide set of architectures. This approach relies on existing software engineering tools from the Java and Scala community, most of them generating OpenCL source code from a high-level abstraction layer.

Dans cette Habilitation à Diriger des Recherches (HDR), nous présentons notre recherche effectuée au cours de ces dernières années dans le domaine du calcul haute-performance. Notre travail a porté essentiellement sur les algorithmes parallèles pour les solveurs d'algèbre linéaire numérique et leur implémentation parallèle dans les bibliothèques logicielles du domaine public. Nous illustrons dans ce manuscrit comment ces calculs peuvent être accélérées en utilisant des algorithmes innovants et être rendus fiables en utilisant des quantités spécifiques de l'analyse d'erreur. Nous expliquons tout d'abord comment les solveurs d'algèbre linéaire numérique peuvent être conçus de façon à exploiter les capacités des calculateurs hétérogènes actuels comprenant des processeurs multicœurs et des GPUs. Nous considérons des algorithmes de factorisation dense pour lesquels nous décrivons la répartition des tâches entre les différentes unités de calcul et son influence en terme de coût des communications. Ces cal- culs peuvent être également rendus plus performants grâce à des algorithmes en précision mixte qui utilisent une précision moindre pour les tâches les plus coûteuses tout en calculant la solution en précision supérieure. Puis nous décrivons notre travail de recherche dans le développement de solveurs d'algèbre linéaire rapides qui utilisent des algorithmes randomisés. La randomisation représente une approche innovante pour accélérer les calculs d'algèbre linéaire et la classe d'algorithmes que nous proposons a l'avantage de réduire la volume de communications dans les factorisations en supprimant complètement la phase de pivotage dans les systèmes linéaires. Les logiciels correspondants on été développés pour architectures multicœurs éventuellement accélérées par des GPUs. Enfin nous proposons des outils qui nous permettent de garantir la qualité de la solution calculée pour les problèmes de moindres carrés sur-déterminés, incluant les moindres carrés totaux. Notre méthode repose sur la dérivation de formules exactes ou d'estimateurs pour le conditionnement de ces problèmes. Nous décrivons les algorithmes et les logiciels qui permettent de calculer ces quantités avec les bibliothèques logicielles parallèles standards. Des pistes de recherche pour les années à venir sont données dans un chapître de conclusion.