Adaptive Mesh Refinement (AMR) is a method which dynamically varies the spatio-temporal resolution of localized mesh regions in numerical simulations, based on the strength of the solution features. Due to high resolution discretization of localized regions of interests into rectangular mesh units called patches, AMR provides low cost of computations and high degree of accuracy. General purpose graphics processing units (GPGPUs) with their support for fine-grained parallelism, offer an attractive option for obtaining high performance for AMR applications. The data parallel computations of the finite difference schemes of AMR can be efficiently performed on GPGPUs. This research deals with challenges and develops techniques for efficient executions of AMR applications with uniform and non-uniform patches on GPUs. In the first part of the thesis, we optimize an AMR model with uniform patches. We have developed strategies for continuous online visualization of time evolving data for AMR applications executed on GPUs. In-situ visualization plays an important role for analyzing the time evolving characteristics of the domain structures. Continuous visualization of the output data for various time steps results in better study of the underlying domain and the model used for simulating the domain. We reorder the meshes for computations on the GPU based on the users input related to the subdomain that he wants to visualize. This makes the data available for visualization at a faster rate. We then perform asynchronous executions of the visualization steps and fix-up operations on the coarse meshes on the CPUs while the GPU advances the solution. By performing experiments on Tesla S1070 and Fermi C2070 clusters, we found that our strategies result in up to 60% improvement in response time and 16% improvement in the rate of visualization of frames over the existing strategy of performing fix-ups and visualization at the end of the time steps. The second part of the thesis deals with adaptive strategies for efficient execution of block structured AMR applications with non-uniform patches on GPUs. Most AMR approaches use patches of uniform sizes over regions of interests. Since this leads to over-refinement, some efforts have focused on forming patches of non-uniform dimensions to improve computational efficiency since the dimensions of a patch can be tuned to the geometry of a region of interest. While effective hybrid execution strategies exist for applications with uniform patches, our work considers efficient execution of non-uniform patches with different workloads. Our techniques include a geometric bin-packing method to load balance GPU computations and reduce thread idling, adaptive determination of amount of work to maximize asynchronism between CPU and GPU executions using a knapsack formulation, and scheduling communications for multi-GPU executions. We test our strategies for synthetic inputs as well as for traces from real applications. Our experiments on Tesla S1070 and Fermi C2070 clusters with both single-GPU and multi-GPU executions show that our strategies result in up to 69% improvement in performance over existing strategies. Our bin-packing based load balancing gives performance gains up to 39%, kernel optimizations give an improvement of up to 20%, and our strategies for adaptive asynchronism between CPU-GPU executions give performance improvements of up to 17% over default static asynchronous executions.

Multichannel acoustic signal processing has undergone major development in recent years due to the increased complexity of current audio processing applications. People want to collaborate through communication with the feeling of being together and sharing the same environment, what is considered as Immersive Audio Schemes. In this phenomenon, several acoustic e ects are involved: 3D spatial sound, room compensation, crosstalk cancelation, sound source localization, among others. However, high computing capacity is required to achieve any of these e ects in a real large-scale system, what represents a considerable limitation for real-time applications. The increase of the computational capacity has been historically linked to the number of transistors in a chip. However, nowadays the improvements in the computational capacity are mainly given by increasing the number of processing units, i.e expanding parallelism in computing. This is the case of the Graphics Processing Units (GPUs), that own now thousands of computing cores. GPUs were traditionally related to graphic or image applications, but new releases in the GPU programming environments, CUDA or OpenCL, allowed that most applications were computationally accelerated in elds beyond graphics. This thesis aims to demonstrate that GPUs are totally valid tools to carry out audio applications that require high computational resources. To this end, di erent applications in the eld of audio processing are studied and performed using GPUs. This manuscript also analyzes and solves possible limitations in each GPU-based implementation both from the acoustic point of view as from the computational point of view. In this document, we have addressed the following problems: Most of audio applications are based on massive ltering. Thus, the rst implementation to undertake is a fundamental operation in the audio processing: the convolution. It has been rst developed as a computational kernel and afterwards used for an application that combines multiples convolutions concurrently: generalized crosstalk cancellation and equalization. The proposed implementation can successfully manage two di erent and common situations: size of bu ers that are much larger than the size of the lters and size of bu ers that are much smaller than the size of the lters. Two spatial audio applications that use the GPU as a co-processor have been developed from the massive multichannel ltering. First application deals with binaural audio. Its main feature is that this application is able to synthesize sound sources in spatial positions that are not included in the database of HRTF and to generate smoothly movements of sound sources. Both features were designed after di erent tests (objective and subjective). The performance regarding number of sound source that could be rendered in real time was assessed on GPUs with di erent GPU architectures. A similar performance is measured in a Wave Field Synthesis system (second spatial audio application) that is composed of 96 loudspeakers. The proposed GPU-based implementation is able to reduce the room e ects during the sound source rendering. A well-known approach for sound source localization in noisy and reverberant environments is also addressed on a multi-GPU system. This is the case of the Steered Response Power with Phase Transform (SRPPHAT) algorithm. Since localization accuracy can be improved by using high-resolution spatial grids and a high number of microphones, accurate acoustic localization systems require high computational power. The solutions implemented in this thesis are evaluated both from localization and from computational performance points of view, taking into account different acoustic environments, and always from a real-time implementation perspective. Finally, This manuscript addresses also massive multichannel ltering when the lters present an In nite Impulse Response (IIR). Two cases are analyzed in this manuscript: 1) IIR lters composed of multiple secondorder sections, and 2) IIR lters that presents an allpass response. Both cases are used to develop and accelerate two di erent applications: 1) to execute multiple Equalizations in a WFS system, and 2) to reduce the dynamic range in an audio signal. / Belloch Rodríguez, JA. (2014). PERFORMANCE IMPROVEMENT OF MULTICHANNEL AUDIO BY GRAPHICS PROCESSING UNITS [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/40651 / TESIS / Premios Extraordinarios de tesis doctorales

This master thesis deals with modern technologies in graphic hardware and using their for general purpose computing. It is primary focused on architecture of unified processors and algorithm implementation via CUDA programming interface. Thesis base is to choose suited algorithm for GPU horsepower demonstration. Main aim of this work is implementation of multiplatform library offering algorithms for discrete volumetric data vectorization. For this purpose was chosen algorithm Marching cubes that is able to find surface of processed object. In created library will be contained algorithm runnable on graphic device and also one runnable on CPU. Finally we compare both variants and discuss their pros and cons.

Tesis por compendio / [ES] En la última década la utilización de la GPGPU (General Purpose computing in Graphics Processing Units; Computación de Propósito General en Unidades de Procesamiento Gráfico) se ha vuelto tremendamente popular en los centros de datos de todo el mundo. Las GPUs (Graphics Processing Units; Unidades de Procesamiento Gráfico) se han establecido como elementos aceleradores de cómputo que son usados junto a las CPUs formando sistemas heterogéneos. La naturaleza masivamente paralela de las GPUs, destinadas tradicionalmente al cómputo de gráficos, permite realizar operaciones numéricas con matrices de datos a gran velocidad debido al gran número de núcleos que integran y al gran ancho de banda de acceso a memoria que poseen. En consecuencia, aplicaciones de todo tipo de campos, tales como química, física, ingeniería, inteligencia artificial, ciencia de materiales, etc. que presentan este tipo de patrones de cómputo se ven beneficiadas, reduciendo drásticamente su tiempo de ejecución. En general, el uso de la aceleración del cómputo en GPUs ha significado un paso adelante y una revolución. Sin embargo, no está exento de problemas, tales como problemas de eficiencia energética, baja utilización de las GPUs, altos costes de adquisición y mantenimiento, etc. En esta tesis pretendemos analizar las principales carencias que presentan estos sistemas heterogéneos y proponer soluciones basadas en el uso de la virtualización remota de GPUs. Para ello hemos utilizado la herramienta rCUDA, desarrollada en la Universitat Politècnica de València, ya que multitud de publicaciones la avalan como el framework de virtualización remota de GPUs más avanzado de la actualidad. Los resutados obtenidos en esta tesis muestran que el uso de rCUDA en entornos de Cloud Computing incrementa el grado de libertad del sistema, ya que permite crear instancias virtuales de las GPUs físicas totalmente a medida de las necesidades de cada una de las máquinas virtuales. En entornos HPC (High Performance Computing; Computación de Altas Prestaciones), rCUDA también proporciona un mayor grado de flexibilidad de uso de las GPUs de todo el clúster de cómputo, ya que permite desacoplar totalmente la parte CPU de la parte GPU de las aplicaciones. Además, las GPUs pueden estar en cualquier nodo del clúster, independientemente del nodo en el que se está ejecutando la parte CPU de la aplicación. En general, tanto para Cloud Computing como en el caso de HPC, este mayor grado de flexibilidad se traduce en un aumento hasta 2x de la productividad de todo el sistema al mismo tiempo que se reduce el consumo energético en un 15%. Finalmente, también hemos desarrollado un mecanismo de migración de trabajos de la parte GPU de las aplicaciones que ha sido integrado dentro del framework rCUDA. Este mecanismo de migración ha sido evaluado y los resultados muestran claramente que, a cambio de una pequeña sobrecarga, alrededor de 400 milisegundos, en el tiempo de ejecución de las aplicaciones, es una potente herramienta con la que, de nuevo, aumentar la productividad y reducir el gasto energético del sistema. En resumen, en esta tesis se analizan los principales problemas derivados del uso de las GPUs como aceleradores de cómputo, tanto en entornos HPC como de Cloud Computing, y se demuestra cómo a través del uso del framework rCUDA, estos problemas pueden solucionarse. Además se desarrolla un potente mecanismo de migración de trabajos GPU, que integrado dentro del framework rCUDA, se convierte en una herramienta clave para los futuros planificadores de trabajos en clusters heterogéneos. / [CA] En l'última dècada la utilització de la GPGPU(General Purpose computing in Graphics Processing Units; Computació de Propòsit General en Unitats de Processament Gràfic) s'ha tornat extremadament popular en els centres de dades de tot el món. Les GPUs (Graphics Processing Units; Unitats de Processament Gràfic) s'han establert com a elements acceleradors de còmput que s'utilitzen al costat de les CPUs formant sistemes heterogenis. La naturalesa massivament paral·lela de les GPUs, destinades tradicionalment al còmput de gràfics, permet realitzar operacions numèriques amb matrius de dades a gran velocitat degut al gran nombre de nuclis que integren i al gran ample de banda d'accés a memòria que posseeixen. En conseqüència, les aplicacions de tot tipus de camps, com ara química, física, enginyeria, intel·ligència artificial, ciència de materials, etc. que presenten aquest tipus de patrons de còmput es veuen beneficiades reduint dràsticament el seu temps d'execució. En general, l'ús de l'acceleració del còmput en GPUs ha significat un pas endavant i una revolució, però no està exempt de problemes, com ara poden ser problemes d'eficiència energètica, baixa utilització de les GPUs, alts costos d'adquisició i manteniment, etc. En aquesta tesi pretenem analitzar les principals mancances que presenten aquests sistemes heterogenis i proposar solucions basades en l'ús de la virtualització remota de GPUs. Per a això hem utilitzat l'eina rCUDA, desenvolupada a la Universitat Politècnica de València, ja que multitud de publicacions l'avalen com el framework de virtualització remota de GPUs més avançat de l'actualitat. Els resultats obtinguts en aquesta tesi mostren que l'ús de rCUDA en entorns de Cloud Computing incrementa el grau de llibertat del sistema, ja que permet crear instàncies virtuals de les GPUs físiques totalment a mida de les necessitats de cadascuna de les màquines virtuals. En entorns HPC (High Performance Computing; Computació d'Altes Prestacions), rCUDA també proporciona un major grau de flexibilitat en l'ús de les GPUs de tot el clúster de còmput, ja que permet desacoblar totalment la part CPU de la part GPU de les aplicacions. A més, les GPUs poden estar en qualsevol node del clúster, sense importar el node en el qual s'està executant la part CPU de l'aplicació. En general, tant per a Cloud Computing com en el cas del HPC, aquest major grau de flexibilitat es tradueix en un augment fins 2x de la productivitat de tot el sistema al mateix temps que es redueix el consum energètic en aproximadament un 15%. Finalment, també hem desenvolupat un mecanisme de migració de treballs de la part GPU de les aplicacions que ha estat integrat dins del framework rCUDA. Aquest mecanisme de migració ha estat avaluat i els resultats mostren clarament que, a canvi d'una petita sobrecàrrega, al voltant de 400 mil·lisegons, en el temps d'execució de les aplicacions, és una potent eina amb la qual, de nou, augmentar la productivitat i reduir la despesa energètica de sistema. En resum, en aquesta tesi s'analitzen els principals problemes derivats de l'ús de les GPUs com acceleradors de còmput, tant en entorns HPC com de Cloud Computing, i es demostra com a través de l'ús del framework rCUDA, aquests problemes poden solucionar-se. A més es desenvolupa un potent mecanisme de migració de treballs GPU, que integrat dins del framework rCUDA, esdevé una eina clau per als futurs planificadors de treballs en clústers heterogenis. / [EN] In the last decade the use of GPGPU (General Purpose computing in Graphics Processing Units) has become extremely popular in data centers around the world. GPUs (Graphics Processing Units) have been established as computational accelerators that are used alongside CPUs to form heterogeneous systems. The massively parallel nature of GPUs, traditionally intended for graphics computing, allows to perform numerical operations with data arrays at high speed. This is achieved thanks to the large number of cores GPUs integrate and the large bandwidth of memory access. Consequently, applications of all kinds of fields, such as chemistry, physics, engineering, artificial intelligence, materials science, and so on, presenting this type of computational patterns are benefited by drastically reducing their execution time. In general, the use of computing acceleration provided by GPUs has meant a step forward and a revolution, but it is not without problems, such as energy efficiency problems, low utilization of GPUs, high acquisition and maintenance costs, etc. In this PhD thesis we aim to analyze the main shortcomings of these heterogeneous systems and propose solutions based on the use of remote GPU virtualization. To that end, we have used the rCUDA middleware, developed at Universitat Politècnica de València. Many publications support rCUDA as the most advanced remote GPU virtualization framework nowadays. The results obtained in this PhD thesis show that the use of rCUDA in Cloud Computing environments increases the degree of freedom of the system, as it allows to create virtual instances of the physical GPUs fully tailored to the needs of each of the virtual machines. In HPC (High Performance Computing) environments, rCUDA also provides a greater degree of flexibility in the use of GPUs throughout the computing cluster, as it allows the CPU part to be completely decoupled from the GPU part of the applications. In addition, GPUs can be on any node in the cluster, regardless of the node on which the CPU part of the application is running. In general, both for Cloud Computing and in the case of HPC, this greater degree of flexibility translates into an up to 2x increase in system-wide throughput while reducing energy consumption by approximately 15%. Finally, we have also developed a job migration mechanism for the GPU part of applications that has been integrated within the rCUDA middleware. This migration mechanism has been evaluated and the results clearly show that, in exchange for a small overhead of about 400 milliseconds in the execution time of the applications, it is a powerful tool with which, again, we can increase productivity and reduce energy foot print of the computing system. In summary, this PhD thesis analyzes the main problems arising from the use of GPUs as computing accelerators, both in HPC and Cloud Computing environments, and demonstrates how thanks to the use of the rCUDA middleware these problems can be addressed. In addition, a powerful GPU job migration mechanism is being developed, which, integrated within the rCUDA framework, becomes a key tool for future job schedulers in heterogeneous clusters. / This work jointly supported by the Fundación Séneca (Agencia Regional de Ciencia y Tecnología, Región de Murcia) under grants (20524/PDC/18, 20813/PI/18 and 20988/PI/18) and by the Spanish MEC and European Commission FEDER under grants TIN2015-66972-C5-3-R, TIN2016-78799-P and CTQ2017-87974-R (AEI/FEDER, UE). We also thank NVIDIA for hardware donation under GPU Educational Center 2014-2016 and Research Center 2015-2016. The authors thankfully acknowledge the computer resources at CTE-POWER and the technical support provided by Barcelona Supercomputing Center - Centro Nacional de Supercomputación (RES-BCV-2018-3-0008). Furthermore, researchers from Universitat Politècnica de València are supported by the Generalitat Valenciana under Grant PROMETEO/2017/077. Authors are also grateful for the generous support provided by Mellanox Technologies Inc. Prof. Pradipta Purkayastha, from Department of Chemical Sciences, Indian Institute of Science Education and Research (IISER) Kolkata, is acknowledged for kindly providing the initial ligand and DNA structures. / Prades Gasulla, J. (2021). Improving Performance and Energy Efficiency of Heterogeneous Systems with rCUDA [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/168081 / TESIS / Compendio

Massive graph visualization in an immersive environment, such as virtual reality (VR) or Augmented Reality (AR), has the possibility to improve users’ understanding when exploring data in new ways. To make the most of a visualization, such as this, requires interactive components that are fast enough to accommodate interactivity. By rendering the edges of the graph as shaded lines that imitate three‑dimensional (3D) lines or tubes, one can circumvent technical limitations. This method works well enough when using traditional two‑dimensional (2D) monitors, but representing tubes as flat lines in a virtual environment (VE) makes for a less immersive user experience as opposed to visualizing true 3D geometry. In order to accommodate for these requirements i.e., speed and visual fidelity, we need a time efficient way of producing tubular meshes. This thesis project explores how one can generate tubular geometry utilizing compute shaders in the modern game engine, Unreal Engine (UE). Exploiting the parallel computing power of the graphical processing unit (GPU) we use compute shaders to generate a tubular mesh following a predetermined path. The result from the project is an open source plugin for UE, able to generate tubular geometry at rapid rates. While not giving any major advantages when generating smaller models, comparing it to a sequential implementation, the compute shader implementation create and render models > 40× faster when generating 106 tube segments. A secondary effect of generating most of the data on the GPU, is that we avoid bottlenecks that can occur when surpassing the bandwidth of the central processing unit (CPU) to GPU data transfer. Using this tool researches can more easily explore information visualization in a VE. Furthermore, this thesis promotes extended development of mesh generation, using compute shaders in UE. / Att visualisera stora grafer i en immersiv miljö, såsom VR eller AR, kan förbättra en användares förståelse när de utforskar data på nya sätt. För att få ut det mesta av denna typen av visualiseringar krävs interaktiva komponenter som är tillräckligt snabba för att tillgodose interaktivitet. Genom att visa de linjer, som binder samman en grafs noder, som plana linjer som imiterar 3Dlinjer eller rör, kan man undvika att slå i det tak som tekniska begränsningar medför. Denna metoden är acceptabel vid användning av traditionella 2Dskärmar, men att representera rör som plana linjer i VE ger en mindre immersiv användarupplevelse, i kontrast till att visualisera sann 3D -geometri. För att tillgodose dessa krav dvs, tidseffektivitet och visuella kvaliteter, behöver vi ett effektivt sätt att producera 3D-linjer. Denna uppsats undersöker hur man kan generera rörformad geometri med hjälp av compute shaders i den moderna spelmotorn Unreal Engine (UE). Genom att använda compute shaders kan vi utnyttja den parallella beräkningskraften hos en GPU, kan vi generera ett rörformat mesh som följer en förutbestämd bana. Resultatet från projektet är ett open source-plugin för UE, som kan generera rörformad geometri i höga hastigheter. Även om det inte kan visas ge några större fördelar när man genererar mindre modeller, jämfört med en sekventiell implementering, skapar och renderar implementeringen av compute Shaders modeller > 40× snabbare, när de genererar 106 rörsegment. I och med att den större delen av datan skapas på GPU kan vi också undvika den flaskhals som kan uppstå när vi överskrider bandbredden mellan CPU och GPU. Med hjälp av verktyget som skapats i samband med denna uppsats kan människor lättare utforska informationsvisualisering i VE. Dessutom främjar denna uppsats utökad utveckling av mesh-generering med hjälp av compute shaders i UE.

In computational fluid dynamics, we often need to solve complex problems with high precision and efficiency. We propose a three-pronged approach to attain this goal. First, we use the discontinuous Galerkin spectral element method (DG-SEM) for its high accuracy. Second, we use graphics processing units (GPUs) to perform our computations to exploit available parallel computing power. Third, we implement a parallel adaptive mesh refinement (AMR) algorithm to efficiently use our computing power where it is most needed. We present a GPU DG-SEM solver with AMR and dynamic load balancing for the 2D wave equation. The DG-SEM is a higher-order method that splits a domain into elements and represents the solution within these elements as a truncated series of orthogonal polynomials. This approach combines the geometric flexibility of finite-element methods with the exponential convergence of spectral methods. GPUs provide a massively parallel architecture, achieving a higher throughput than traditional CPUs. They are relatively new as a platform in the scientific community, therefore most algorithms need to be adapted to that new architecture. We perform most of our computations in parallel on multiple GPUs. AMR selectively refines elements in the domain where the error is estimated to be higher than a prescribed tolerance, via two mechanisms: p-refinement increases the polynomial order within elements, and h-refinement splits elements into several smaller ones. This provides a higher accuracy in important flow regions and increases capabilities of modeling complex flows, while saving computing power in other parts of the domain. We use the mortar element method to retain the exponential convergence of high-order methods at the non-conforming interfaces created by AMR. We implement a parallel dynamic load balancing algorithm to even out the load imbalance caused by solving problems in parallel over multiple GPUs with AMR. We implement a space-filling curve-based repartitioning algorithm which ensures good locality and small interfaces. While the intense calculations of the high order approach suit the GPU architecture, programming of the highly dynamic adaptive algorithm on GPUs is the most challenging aspect of this work. The resulting solver is tested on up to 64 GPUs on HPC platforms, where it shows good strong and weak scaling characteristics. Several example problems of increasing complexity are performed, showing a reduction in computation time of up to 3× on GPUs vs CPUs, depending on the loading of the GPUs and other user-defined choices of parameters. AMR is shown to improve computation times by an order of magnitude or more.

No description available.

Ces dernières années avec les progrès matériels, les dimensions et le contenu des images acquises se sont complexifiés de manière notable. Egalement, le différentiel de performance entre les architectures classiques mono-processeur et parallèles est passé résolument en faveur de ces dernières. Pourtant, les manières de programmer sont restées largement les mêmes, instituant un manque criant de performance même sur ces architectures. Dans cette thèse, nous explorons en détails un algorithme particulier, les flots maximaux continus. Nous explicitons pourquoi cet algorithme est important et utile, et nous proposons plusieurs implémentations sur diverses architectures, du mono-processeur à l'architecture SMP et NUMA, ainsi que sur les architectures massivement parallèles des GPGPU. Nous explorons aussi des applications et nous évaluons ses performances sur des images de grande taille en science des matériaux et en biologie à l'échelle nano / In recent years, with the advance of computing equipment and image acquisition techniques, the sizes, dimensions and content of acquired images have increased considerably. Unfortunately as time passes there is a steadily increasing gap between the classical and parallel programming paradigms and their actual performance on modern computer hardware. In this thesis we consider in depth one particular algorithm, the continuous maximum flow computation. We review in detail why this algorithm is useful and interesting, and we propose efficient and portable implementations on various architectures. We also examine how it performs in the terms of segmentation quality on some recent problems of materials science and nano-scale biology

In this project, various computational energy landscape methods were accelerated using graphics processing units (GPUs). Basin-hopping global optimisation was treated using a version of the limited-memory BFGS algorithm adapted for CUDA, in combination with GPU-acceleration of the potential calculation. The Lennard-Jones potential was implemented using CUDA, and an interface to the GPU-accelerated AMBER potential was constructed. These results were then extended to form the basis of a GPU-accelerated version of hybrid eigenvector-following. The doubly-nudged elastic band method was also accelerated using an interface to the potential calculation on GPU. Additionally, a local rigid body framework was adapted for GPU hardware. Tests were performed for eight biomolecules represented using the AMBER potential, ranging in size from 81 to 22\,811 atoms, and the effects of minimiser history size and local rigidification on the overall efficiency were analysed. Improvements relative to CPU performance of up to two orders of magnitude were obtained for the largest systems. These methods have been successfully applied to both biological systems and atomic clusters. An existing interface between a code for free energy basin-hopping and the SuiteSparse package for sparse Cholesky factorisation was refined, validated and tested. Tests were performed for both Lennard-Jones clusters and selected biomolecules represented using the AMBER potential. Significant acceleration of the vibrational frequency calculations was achieved, with negligible loss of accuracy, relative to the standard diagonalisation procedure. For the larger systems, exploiting sparsity reduces the computational cost by factors of 10 to 30. The acceleration of these computational energy landscape methods opens up the possibility of investigating much larger and more complex systems than previously accessible. A wide array of new applications are now computationally feasible.

Le travail présenté dans ce mémoire étudie et propose des modèles de calcul parallèles de type cellulaire pour traiter différents problèmes d’optimisation NP-durs définis dans l’espace euclidien, et leur implantation sur des processeurs graphiques multi-fonction (Graphics Processing Unit; GPU). Le but est de pouvoir traiter des problèmes de grande taille tout en permettant des facteurs d’accélération substantiels à l’aide du parallélisme massif. Les champs d’application visés concernent les systèmes embarqués pour la stéréovision de même que les problèmes de transports définis dans le plan, tels que les problèmes de tournées de véhicules. La principale caractéristique du modèle cellulaire est qu’il est fondé sur une décomposition du plan en un nombre approprié de cellules, chacune comportant une part constante de la donnée, et chacune correspondant à une unité de calcul (processus). Ainsi, le nombre de processus parallèles et la taille mémoire nécessaire sont en relation linéaire avec la taille du problème d’optimisation, ce qui permet de traiter des instances de très grandes tailles.L’efficacité des modèles cellulaires proposés a été testée sur plateforme parallèle GPU sur quatre applications. La première application est un problème d’appariement d’images stéréo. Elle concerne la stéréovision couleur. L’entrée du problème est une paire d’images stéréo, et la sortie une carte de disparités représentant les profondeurs dans la scène 3D. Le but est de comparer des méthodes d’appariement local selon l’approche winner-takes-all et appliquées à des paires d’images CFA (color filter array). La deuxième application concerne la recherche d’améliorations de l’implantation GPU permettant de réaliser un calcul quasi temps-réel de l’appariement. Les troisième et quatrième applications ont trait à l’implantation cellulaire GPU des réseaux neuronaux de type carte auto-organisatrice dans le plan. La troisième application concerne la génération de maillages structurés appliquée aux cartes de disparité afin de produire des représentations compressées des surfaces 3D. Enfin, la quatrième application concerne le traitement d’instances de grandes tailles du problème du voyageur de commerce euclidien comportant jusqu’à 33708 villes.Pour chacune des applications, les implantations GPU permettent une accélération substantielle du calcul par rapport aux versions CPU, pour des tailles croissantes des problèmes et pour une qualité de résultat obtenue similaire ou supérieure. Le facteur d’accélération GPU par rapport à la version CPU est d’environ 20 fois plus vite pour la version GPU sur le traitement des images CFA, cependant que le temps de traitement GPU est d’environ de 0,2s pour une paire d’images de petites tailles de la base Middlebury. L’algorithme amélioré quasi temps-réel nécessite environ 0,017s pour traiter une paire d’images de petites tailles, ce qui correspond aux temps d’exécution parmi les plus rapides de la base Middlebury pour une qualité de résultat modérée. La génération de maillages structurés est évaluée sur la base Middlebury afin de déterminer les facteurs d’accélération et qualité de résultats obtenus. Le facteur d’accélération obtenu pour l’implantation parallèle des cartes auto-organisatrices appliquée au problème du voyageur de commerce et pour l’instance avec 33708 villes est de 30 pour la version parallèle. / The work presented in this PhD studies and proposes cellular computation parallel models able to address different types of NP-hard optimization problems defined in the Euclidean space, and their implementation on the Graphics Processing Unit (GPU) platform. The goal is to allow both dealing with large size problems and provide substantial acceleration factors by massive parallelism. The field of applications concerns vehicle embedded systems for stereovision as well as transportation problems in the plane, as vehicle routing problems. The main characteristic of the cellular model is that it decomposes the plane into an appropriate number of cellular units, each responsible of a constant part of the input data, and such that each cell corresponds to a single processing unit. Hence, the number of processing units and required memory are with linear increasing relationship to the optimization problem size, which makes the model able to deal with very large size problems.The effectiveness of the proposed cellular models has been tested on the GPU parallel platform on four applications. The first application is a stereo-matching problem. It concerns color stereovision. The problem input is a stereo image pair, and the output a disparity map that represents depths in the 3D scene. The goal is to implement and compare GPU/CPU winner-takes-all local dense stereo-matching methods dealing with CFA (color filter array) image pairs. The second application focuses on the possible GPU improvements able to reach near real-time stereo-matching computation. The third and fourth applications deal with a cellular GPU implementation of the self-organizing map neural network in the plane. The third application concerns structured mesh generation according to the disparity map to allow 3D surface compressed representation. Then, the fourth application is to address large size Euclidean traveling salesman problems (TSP) with up to 33708 cities.In all applications, GPU implementations allow substantial acceleration factors over CPU versions, as the problem size increases and for similar or higher quality results. The GPU speedup factor over CPU was of 20 times faster for the CFA image pairs, but GPU computation time is about 0.2s for a small image pair from Middlebury database. The near real-time stereovision algorithm takes about 0.017s for a small image pair, which is one of the fastest records in the Middlebury benchmark with moderate quality. The structured mesh generation is evaluated on Middlebury data set to gauge the GPU acceleration factor and quality obtained. The acceleration factor for the GPU parallel self-organizing map over the CPU version, on the largest TSP problem with 33708 cities, is of 30 times faster.