GPGPU (General purpose computing on graphics processing unit) is quite common in today's modern computer games when doing heavy simulation calculations like game physics or particle systems. GPU programming is not only used in games but also in scientific research when doing heavy calculations on molecular structures and protein folding etc. The reason why you use the GPU for these kinds of tasks is that you can gain an incredible speedup in performance to your application. Previous research shows that particle systems scale very well to the GPU architecture. When simulating very large particle-system on the GPU it can run up to 79 times faster than the CPU. But for some very small particle systems the CPU proved to be faster. This research aims to compare the difference between the GPU and CPU when it comes to simulating many smaller particle-systems and to see what happen to the performance when the particle-systems become smaller and smaller.

Målet med projektet var att undersöka hur skeletal animationer utförs. Ett mål till var att det skulle vara förbestämda animationer. För att kunna ha förberäknade animationer användes autodesk fbx-filer. Skinningen har använt dubbel kvaternioner istället för matriser.Rapporten visar att skeletal animation med dubbel kvaternion skinning teknik kan utföras genom att använda fbx-filer med data som först exporterats till json-format.

Image completion is a process of removing an area from a photograph and replacing it with suitable data. Earlier methods either search for this relevant data within the image itself, or extends the search to some form of additional data, usually some form of database. Methods that search for suitable data within the image itself has problems when no suitable data can be found in the image. Methods that extend their search has in earlier work either used some form of database with labeled images or a massive database with photos from the Internet. For the labels in a database to be useful they typically needs to be entered manually, which is a very time consuming process. Methods that uses databases with millions of images from the Internet has issues with copyrighted images, storage of the photographs and computation time. This work shows that a small database of the user’s own private, or professional, photos can be used to improve the quality of image completions. A photographer today typically take many similar photographs on similar scenes during a photo session. Therefore a smaller number of images are needed to find images that are visually and structurally similar, than when random images downloaded from the internet are used. Thus, this approach gains most of the advantages of using additional data for the image completions, while at the same time minimizing the disadvantages. It gains a better ability to find suitable data without having to process millions of irrelevant photos.

Bio-pharmaceutical industry is a vast growing market and recent recommendations of the Food and Drug Administration have put a large emphasis on the characterization of biological processes and models. As a consequence, there is a high incentive on developing modern sensors in order to more accurately monitor and control processes. In that way, Digital Holographic Microscopy (DHM) presents unique features thanks to the refocusing and quantitative phase contrast imaging capabilities. In this thesis we investigate the usage of DHM to monitor yeast cultures that are often used in both the bio-pharmaceutical and bread industries and lay the basis of a methodological framework for the study of in-line cell cultures in the context of process control. We begin with a description of Digital Holography and the microscopy setup used in the thesis as well as a detailed explanation of the image processing required to extract the holographic data and its implementation on GPU with some speed execution figures given for three popular programming paradigms. We then describe the flow setup used and infer the limitations on the dynamic range of the technique due to both Poisson statistics and overlapping phenomena. Finally, we describe an algorithm that extracts the cells position, count and morphological information such as the size, aspect ratio, circularity and refraction index. Some experimental results are presented for yeasts before drawing a general overview of the technology and its dependencies. We further end with some conclusions concerning the technology and a brief comparison with existing competitors. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

El presente trabajo se inscribe en el campo de la computación paralela y, más en concreto, en el desarrollo y utilización de modelos computacionales en arquitecturas paralelas heterogéneas para la resolución de problemas aplicados. En la tesis abordamos una serie de problemas que están relacionados con la aplicación de la tecnología en el ámbito de las explotaciones agrícolas y comprenden: la representación del relieve, el manejo de información climática como la temperatura, y la gestión de recursos hídricos. El estudio y la solución a estos problemas en el área en la que se han estudiado tienen un amplio impacto económico y medioambiental. Los problemas basan su formulación en un modelo matemático cuya solución es costosa desde el punto de vista computacional, siendo incluso a veces inviable. La tesis consiste en implementar algoritmos paralelos rápidos y eficientes que resuelven el problema matemático asociado a estos problemas en nodos multicore y multi-GPU. También se estudia, propone y aplican técnicas que permiten a las rutinas diseñadas adaptarse automáticamente a las características del sistema paralelo donde van a ser instaladas y ejecutadas con el objeto de obtener la versión más cercana posible a la óptima a un bajo coste. El objetivo es proporcionar un software a los usuarios que sea portable, pero a la vez, capaz de ejecutarse eficientemente en la ordenador donde se esté trabajando, independientemente de las características de la arquitectura y de los conocimientos que el usuario pueda tener sobre dicha arquitectura. / Do Carmo Boratto, M. (2015). Modelos Paralelos para la Resolución de Problemas de Ingeniería Agrícola [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/48529 / TESIS

This thesis deals with Burrows-Wheeler transform (BWT) and possibilities of acceleration of this transform on graphics processing unit (GPU). Methods of compression based on BWT are introduced, as well as software libraries CUDA and OpenCL for writing programs for GPU. Parallel variants of BWT are implemented, as well as following steps necessary for compression, using CUDA library. Amount of compression of used approaches are tested and parallel versions are compared to their sequential counterparts.

Hlavním cílem této práce je navrhnout novu metodu rekonstrukce obrazu z dat fotoakustického snímkování. Fotoakustické snímkování je velmi populární neinvazivní metoda snímkování založená na detekování ultrazvukových vln vyvolaných laserovým paprskem. Proces snímkování generuje velké množství dat, a kvůli tomu je proces rekonstrukce obrazu velmi časově náročný. Táto práce demonstruje proces rekonstrukce obrazu pomocí zpětné projekce, algoritmu který je dostatečně jednoduchý na přizpůsobení moderním architekturám procesorů umožňující různé způsoby optimalizovaného výpočtu. Dvě různé variantu algoritmu byly navrženy: z pohledu pixelu a z pohledu senzoru, který detekuje ultrazvukové vlny. Obě varianty byly implementovány třemi různými způsoby: pomocí vektorového paralelismu, vláknového paralelismu a paralelismu na grafické karetě (GPU). Všechny 3 implementace obou variant algoritmu byly testovány a výsledky byly srovnány s výsledkem rekonstrukce algoritmu reverzního času, přesnějšího ale mnohokrát pomalejšího algoritmu. Výsledky ukázaly, že GPU paralelismus nabízí nejrychlejší výpočet, cca. 200 krát rychlejší než u algoritmu reverzního času, a proto se dá použit i v aplikacích pracující v reálném čase.

The PRINCE algorithm is a faster and more advanced version of a combination attack. Non-distributed password breaking often encounters its limits, and its applicability to real tasks decreases due to the increasing demand for computing resources of the device. The aim of this work is to design a distributed version of the the PRINCE attack as an extension of Fitcrack system, which focuses on distributed password cracking. The proposed design is implemented and integrated into the Fitcrack system. The work examines the PRINCE attack on a set of experiments, which examines the impact of various configuration options. Part of the experimental part is a comparison of the PRINCE attack with the dictionary and combination attack. The purpose of the comparison is to find cases where the PRINCE attack is better than other attacks. Finally, the integrated PRINCE attack solution in the Fitcrack system is compared with the solution implemented in the Hashtopolis system.

This work describes one - way hash functions and cryptographic algorithms . It also describes their implementation regarding DOC, PDF and ZIP files contents encryption . Subsequently , the implementation analyzis is provided . Following next, the brute - force attack procedure levereging GPU is proposed and evaluated.

This master's thesis deals with acceleration of geometrical image transforms using the GPU and NVIDIA (R) CUDA TM architecture. Time critical parts of the code are moved on the GPU and executed in parallel. One of the results is a demonstrational application for performance comparison of both architectures: the CPU, and GPU in combination with the CPU. As a reference implementation, there are used highly optimized routines from the OpenCV library, made by the Intel company.