OpenStack is currently the most popular open source platform for Infrastructure as a Service (IaaS) clouds. OpenStack lets users deploy virtual machines and other instances, which handle different tasks for managing a cloud environment on the fly. A lot of cloud platform offerings, including the Ericsson Cloud System, are based on OpenStack. Despite the popularity of OpenStack, there is currently a limited understanding of how much resource is consumed/needed by components of OpenStack under different operating conditions such as number of compute nodes, number of running VMs, the number of users and the rate of requests to the various services. The master thesis attempts to model the resource demand of the various components of OpenStack in function of different operating condition, identify correlations and evaluate how accurate the predictions are. For this purpose, a physical OpenStack is setup with one strong controller node and eight compute nodes. All the experiments and measurements were on virtual OpenStack components on top of the main physical one. In conclusion, a simple model is generated for idle behavior of OpenStack, starting and stopping a Virtual Machine (VM) API calls which predicts the total CPU utilization based on the number of Compute Nodes and VMs.

Researchers have made significant progress in information extraction from short documents in the last few years, including social media interaction, news articles, and email excerpts. This research aims to extract technical entities like hardware resources, computing platforms, compute time, programming language, and libraries from scholarly research articles. Research articles are generally long documents having both salient as well as non-salient entities. Analyzing the cross-sectional relation, filtering the relevant information, measuring the saliency of mentioned entities, and extracting novel entities are some of the technical challenges involved in this research. This work presents a detailed study about the performance, effectiveness, and scalability of rule-based weakly supervised algorithms. We also develop an automated end-to-end Research Entity and Relationship Extractor (E2R Extractor). Additionally, we perform a comprehensive study about the effectiveness of existing deep learning-based information extraction tools like Dygie, Dygie++, SciREX. The research also contributes a dataset containing novel entities annotated in BILUO format and represents the baseline results using the E2R extractor on the proposed dataset. The results indicate that the E2R extractor successfully extracts salient entities from research articles. / Master of Science / Information extraction is a process of automatically extracting meaningful information from unstructured text such as articles, news feeds and presenting it in a structured format. Researchers have made significant progress in this domain over the past few years. However, their work primarily focuses on short documents such as social media interactions, news articles, email excerpts, and not on long documents such as scholarly articles and research papers. Long documents contain a lot of redundant data, so filtering and extracting meaningful information is quite challenging. This work focuses on extracting entities such as hardware resources, compute platforms, and programming languages used in scholarly articles. We present a deep learning-based model to extract such entities from research articles and research papers. We evaluate the performance of our deep learning model against simple rule-based algorithms and other state-of-the-art models for extracting the desired entities. Our work also contributes a labeled dataset containing the entities mentioned above and results obtained on this dataset using our deep learning model.

Os métodos numéricos convencionais, baseados em malhas, têm sido amplamente aplicados na resolução de problemas da Dinâmica dos Fluidos Computacional. Entretanto, em problemas de escoamento de fluidos que envolvem superfícies livres, grandes explosões, grandes deformações, descontinuidades, ondas de choque etc., estes métodos podem apresentar algumas dificuldades práticas quando da resolução destes problemas. Como uma alternativa viável, existem os métodos de partículas livre de malhas. Neste trabalho é feita uma introdução ao método Lagrangeano de partículas, livre de malhas, Smoothed Particle Hydrodynamics (SPH) voltado para a simulação numérica de escoamentos de fluidos newtonianos compressíveis e quase-incompressíveis. Dois códigos numéricos foram desenvolvidos, uma versão serial e outra em paralelo, empregando a linguagem de programação C/C++ e a Compute Unified Device Architecture (CUDA), que possibilita o processamento em paralelo empregando os núcleos das Graphics Processing Units (GPUs) das placas de vídeo da NVIDIA Corporation. Os resultados numéricos foram validados e a eficiência computacional avaliada considerandose a resolução dos problemas unidimensionais Shock Tube e Blast Wave e bidimensional da Cavidade (Shear Driven Cavity Problem). / The conventional mesh-based numerical methods have been widely applied to solving problems in Computational Fluid Dynamics. However, in problems involving fluid flow free surfaces, large explosions, large deformations, discontinuities, shock waves etc. these methods suffer from some inherent difficulties which limit their applications to solving these problems. Meshfree particle methods have emerged as an alternative to the conventional grid-based methods. This work introduces the Smoothed Particle Hydrodynamics (SPH), a meshfree Lagrangian particle method to solve compressible flows. Two numerical codes have been developed, serial and parallel versions, using the Programming Language C/C++ and Compute Unified Device Architecture (CUDA). CUDA is NVIDIAs parallel computing architecture that enables dramatic increasing in computing performance by harnessing the power of the Graphics Processing Units (GPUs). The numerical results were validated and the speedup evaluated for the Shock Tube and Blast Wave one-dimensional problems and Shear Driven Cavity Problem.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Neste trabalho, foi desenvolvido um simulador numérico baseado no método livre de malhas Smoothed Particle Hydrodynamics (SPH) para a resolução de escoamentos de fluidos newtonianos incompressíveis. Diferentemente da maioria das versões existentes deste método, o código numérico faz uso de uma técnica iterativa na determinação do campo de pressões. Este procedimento emprega a forma diferencial de uma equação de estado para um fluido compressível e a equação da continuidade a fim de que a correção da pressão seja determinada. Uma versão paralelizada do simulador numérico foi implementada usando a linguagem de programação C/C++ e a Compute Unified Device Architecture (CUDA) da NVIDIA Corporation. Foram simulados três problemas, o problema unidimensional do escoamento de Couette e os problemas bidimensionais do escoamento no interior de uma Cavidade (Shear Driven Cavity Problem) e da Quebra de Barragem (Dambreak). / In this work a numerical simulator was developed based on the mesh-free Smoothed Particle Hydrodynamics (SPH) method to solve incompressible newtonian fluid flows. Unlike most existing versions of this method, the numerical code uses an iterative technique in the pressure field determination. This approach employs a differential state equation for a compressible fluid and the continuity equation to calculate the pressure correction. A parallel version of the numerical code was implemented using the Programming Language C/C++ and Compute Unified Device Architecture (CUDA) from the NVIDIA Corporation. The numerical results were validated and the speed-up evaluated for an one-dimensional Couette flow and two-dimensional Shear Driven Cavity and Dambreak problems.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Neste trabalho, foi desenvolvido um simulador numérico baseado no método livre de malhas Smoothed Particle Hydrodynamics (SPH) para a resolução de escoamentos de fluidos newtonianos incompressíveis. Diferentemente da maioria das versões existentes deste método, o código numérico faz uso de uma técnica iterativa na determinação do campo de pressões. Este procedimento emprega a forma diferencial de uma equação de estado para um fluido compressível e a equação da continuidade a fim de que a correção da pressão seja determinada. Uma versão paralelizada do simulador numérico foi implementada usando a linguagem de programação C/C++ e a Compute Unified Device Architecture (CUDA) da NVIDIA Corporation. Foram simulados três problemas, o problema unidimensional do escoamento de Couette e os problemas bidimensionais do escoamento no interior de uma Cavidade (Shear Driven Cavity Problem) e da Quebra de Barragem (Dambreak). / In this work a numerical simulator was developed based on the mesh-free Smoothed Particle Hydrodynamics (SPH) method to solve incompressible newtonian fluid flows. Unlike most existing versions of this method, the numerical code uses an iterative technique in the pressure field determination. This approach employs a differential state equation for a compressible fluid and the continuity equation to calculate the pressure correction. A parallel version of the numerical code was implemented using the Programming Language C/C++ and Compute Unified Device Architecture (CUDA) from the NVIDIA Corporation. The numerical results were validated and the speed-up evaluated for an one-dimensional Couette flow and two-dimensional Shear Driven Cavity and Dambreak problems.

Made available in DSpace on 2015-05-14T12:36:40Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 2336902 bytes, checksum: 41580878008a0f84da693637a48ceb33 (MD5) Previous issue date: 2013-08-08 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The present work deals with the Continuous Global Optimization Problem, in its minimization form, by testing two approaches for the Continuous Greedy Randomized Adaptive Search Procedure (C-GRASP). The development of the first method - sequential and hybrid - comes from the deficiency of current approaches to provide a good neighborhood space exploration. Being constructed from the combination of two meta-heuristics, standard C-GRASP and Continuous General Variable Neighborhood Search (C-GVNS), as a strategy to achieving symmetric trades of neighborhood structures, it performed efficiently in the computational tests that were taken. The second procedure arises from the large consume of time when using high dimension functions with the standard C-GRASP construction procedure. As the optimization problems have a high dimensionality increase, it s preferable to have two parallel versions of the optimization method in order to handle bigger problems. Thus, for this new procedure developed, it was used the Compute Unified Device Architecture (CUDA), which provided promising acceleration regarding the processing time, based on the experiments performed. / O presente trabalho aborda o Problema de Otimização Global Contínua, em sua forma de minimização, através de duas abordagens para o procedimento Continuous Greedy Randomized Adaptive Search Procedure (C-GRASP). A elaboração do primeiro método, sequencial e híbrido, parte da deficiência presente nas abordagens atuais, em promover boa exploração no espaço de vizinhança. Sendo constituída da combinação de duas meta-heurísticas, C-GRASP padrão e Continuous General Variable Neighborhood Search (C-GVNS). Como estratégia para a realização de trocas sistemática de estruturas de vizinhanças, mostrou-se eficiente aos testes computacionais realizados. O segundo procedimento elaborado parte do grande consumo de tempo ao utilizar funções com alta dimensão, pelo procedimento de construção do método C-GRASP padrão. Como os problemas de otimização possuem crescimento elevado de dimensionalidade, é desejável ter versões paralelas do método de otimização para lidar com os problemas maiores. Desta forma, para o novo procedimento elaborado foi empregado a plataforma de computação paralela Compute Unified Device Architecture (CUDA), que, conforme verificado nos experimentos realizados, promoveu promissora aceleração quanto ao tempo de processamento.

Os métodos numéricos convencionais, baseados em malhas, têm sido amplamente aplicados na resolução de problemas da Dinâmica dos Fluidos Computacional. Entretanto, em problemas de escoamento de fluidos que envolvem superfícies livres, grandes explosões, grandes deformações, descontinuidades, ondas de choque etc., estes métodos podem apresentar algumas dificuldades práticas quando da resolução destes problemas. Como uma alternativa viável, existem os métodos de partículas livre de malhas. Neste trabalho é feita uma introdução ao método Lagrangeano de partículas, livre de malhas, Smoothed Particle Hydrodynamics (SPH) voltado para a simulação numérica de escoamentos de fluidos newtonianos compressíveis e quase-incompressíveis. Dois códigos numéricos foram desenvolvidos, uma versão serial e outra em paralelo, empregando a linguagem de programação C/C++ e a Compute Unified Device Architecture (CUDA), que possibilita o processamento em paralelo empregando os núcleos das Graphics Processing Units (GPUs) das placas de vídeo da NVIDIA Corporation. Os resultados numéricos foram validados e a eficiência computacional avaliada considerandose a resolução dos problemas unidimensionais Shock Tube e Blast Wave e bidimensional da Cavidade (Shear Driven Cavity Problem). / The conventional mesh-based numerical methods have been widely applied to solving problems in Computational Fluid Dynamics. However, in problems involving fluid flow free surfaces, large explosions, large deformations, discontinuities, shock waves etc. these methods suffer from some inherent difficulties which limit their applications to solving these problems. Meshfree particle methods have emerged as an alternative to the conventional grid-based methods. This work introduces the Smoothed Particle Hydrodynamics (SPH), a meshfree Lagrangian particle method to solve compressible flows. Two numerical codes have been developed, serial and parallel versions, using the Programming Language C/C++ and Compute Unified Device Architecture (CUDA). CUDA is NVIDIAs parallel computing architecture that enables dramatic increasing in computing performance by harnessing the power of the Graphics Processing Units (GPUs). The numerical results were validated and the speedup evaluated for the Shock Tube and Blast Wave one-dimensional problems and Shear Driven Cavity Problem.

Made available in DSpace on 2016-03-15T19:38:03Z (GMT). No. of bitstreams: 1 TIAGO UNGARO BARDELLA.pdf: 2553991 bytes, checksum: f8e6ba33f7c930ee81f6b64116f495ff (MD5) Previous issue date: 2015-10-19 / The history of digital games shows, since the beginning, games which uses many types of enemy models to confront and many types of characters to control, like Real-Time Strategy games, for example. These huge amount of models into an important scene are called crowds. The crowds needs a high computer performance and specific algorithms in their interaction control to avoid immersion loss into a game by problems which may happen if the crowds are not treated accordingly. With the popularization of graphic board languages like NVIDIA CUDA, new algorithms were created to easily increase the performance of crowds in digital games and their overwhelming superiority compared to the methods used in linear programming were proved in many researches. The goal of this work is to use these GPU techniques as base to implement a new API using CUDA language that will present better performance and simplicity compared to the others algorithms on the area of crowds in digital games. After the project conclusion, the created API turned easier the crowd treatment to digital game developers using Unity3D integrated with API TBX, that now only need to include a DLL in the project instead creating na algorithm for crowd treatment from the beginning, which takes a huge amount of time from development. / O histórico dos jogos digitais apresenta, desde seu princípio, jogos que utilizam diversos modelos de inimigos para enfrentar ou diversos modelos de personagens para controlar, como os jogos Real-Time Strategy por exemplo. Essas grandes quantidades de modelos que compõem uma cena importante são chamadas de multidões. As multidões necessitam de um alto poder computacional e algoritmos específicos para seu tratamento para evitar a perda de imersão dentro de um jogo pelos problemas que podem acontecer caso as multidões não sejam tratadas adequadamente. Com o surgimento de linguagens de placas gráficas como a NVIDIA CUDA, novos algoritmos foram criados para melhor trabalhar com o desempenho de multidões em jogos digitais e sua superioridade em comparação com os métodos utilizados em programação sequencial foi comprovada em diversos estudos. O objetivo deste trabalho é se basear nestas técnicas de GPU para implementar uma nova API usando tecnologia CUDA que visa melhorar os algoritmos existentes para tratamento de multidões em jogos digitais em termos de desempenho e simplicidade de implementação. Com a conclusão do projeto, a API criada facilitou o tratamento de multidões para desenvolvedores de jogos digitais com a game engine Unity3D integrada com a API TBX de simulação de multidões, que agora apenas precisam incluir uma DLL em seu projeto ao invés de criar um algoritmo próprio de tratamento de multidões do início, o que demanda tempo de desenvolvimento.

In this thesis, we investigate physical-layer network coding in an L × M × K relay network, where L source nodes want to transmit messages to K sink nodes via M relay nodes. We focus on the information processing at the relay nodes and the compute-and-forward framework. Nested lattice codes are used, which have the property that every linear combination of codewords is a valid codeword. This property is essential for physical-layer network coding. Because the actual network coding occurs on the physical layer, the network coding coefficients are determined by the channel realizations. Finding the optimal network coding coefficients for given channel realizations is a non-trivial optimization problem. In this thesis, we provide an algorithm to find network coding coefficients that result in the highest data rate at a chosen relay. The solution of this optimization problem is only locally optimal, i.e., it is optimal for a particular relay. If we consider a multi-hop network, each potential receiver must get enough linear independent combinations to be able to decode the individual messages. If this is not the case, outage occurs, which results in data loss. In this thesis, we propose a new strategy for choosing the network coding coefficients locally at the relays without solving the optimization problem globally. We thereby reduce the solution space for the relays such that linear independence between their decoded linear combinations is guaranteed. Further, we discuss the influence of spatial correlation on the optimization problem. Having solved the optimization problem, we combine physical-layer network coding with physical-layer secrecy. This allows us to propose a coding scheme to exploit untrusted relays in multi-user relay networks. We show that physical-layer network coding, especially compute-and-forward, is a key technology for simultaneous and secure communication of several users over an untrusted relay. First, we derive the achievable secrecy rate for the two-way relay channel. Then, we enhance this scenario to a multi-way relay channel with multiple antennas. We describe our implementation of the compute-and-forward framework with software-defined radio and demonstrate the practical feasibility. We show that it is possible to use the framework in real-life scenarios and demonstrate a transmission from two users to a relay. We gain valuable insights into a real transmission using the compute-and-forward framework. We discuss possible improvements of the current implementation and point out further work. / In dieser Arbeit untersuchen wir Netzwerkcodierung auf der Übertragungsschicht in einem Relay-Netzwerk, in dem L Quellen-Knoten Nachrichten zu K Senken-Knoten über M Relay-Knoten senden wollen. Der Fokus dieser Arbeit liegt auf der Informationsverarbeitung an den Relay-Knoten und dem Compute-and-Forward Framework. Es werden Nested Lattice Codes eingesetzt, welche die Eigenschaft besitzen, dass jede Linearkombination zweier Codewörter wieder ein gültiges Codewort ergibt. Dies ist eine Eigenschaft, die für die Netzwerkcodierung von entscheidender Bedeutung ist. Da die eigentliche Netzwerkcodierung auf der Übertragungsschicht stattfindet, werden die Netzwerkcodierungskoeffizienten von den Kanalrealisierungen bestimmt. Das Finden der optimalen Koeffizienten für gegebene Kanalrealisierungen ist ein nicht-triviales Optimierungsproblem. Wir schlagen in dieser Arbeit einen Algorithmus vor, welcher Netzwerkcodierungskoeffizienten findet, die in der höchsten Übertragungsrate an einem gewählten Relay resultieren. Die Lösung dieses Optimierungsproblems ist zunächst nur lokal, d. h. für dieses Relay, optimal. An jedem potentiellen Empfänger müssen ausreichend unabhängige Linearkombinationen vorhanden sein, um die einzelnen Nachrichten decodieren zu können. Ist dies nicht der Fall, kommt es zu Datenverlusten. Um dieses Problem zu umgehen, ohne dabei das Optimierungsproblem global lösen zu müssen, schlagen wir eine neue Strategie vor, welche den Lösungsraum an einem Relay soweit einschränkt, dass lineare Unabhängigkeit zwischen den decodierten Linearkombinationen an den Relays garantiert ist. Außerdem diskutieren wir den Einfluss von räumlicher Korrelation auf das Optimierungsproblem. Wir kombinieren die Netzwerkcodierung mit dem Konzept von Sicherheit auf der Übertragungsschicht, um ein Übertragungsschema zu entwickeln, welches es ermöglicht, mit Hilfe nicht-vertrauenswürdiger Relays zu kommunizieren. Wir zeigen, dass Compute-and-Forward ein wesentlicher Baustein ist, um solch eine sichere und simultane Übertragung mehrerer Nutzer zu gewährleisten. Wir starten mit dem einfachen Fall eines Relay-Kanals mit zwei Nutzern und erweitern dieses Szenario auf einen Relay-Kanal mit mehreren Nutzern und mehreren Antennen. Die Arbeit wird abgerundet, indem wir eine Implementierung des Compute-and-Forward Frameworks mit Software-Defined Radio demonstrieren. Wir zeigen am Beispiel von zwei Nutzern und einem Relay, dass sich das Framework eignet, um in realen Szenarien eingesetzt zu werden. Wir diskutieren mögliche Verbesserungen und zeigen Richtungen für weitere Forschungsarbeit auf.

Les futures générations de réseaux sans fil posent beaucoup de défis pour la communauté de recherche. Notamment, ces réseaux doivent être en mesure de répondre, avec une certaine qualité de service, aux demandes d'un nombre important de personnes et d'objets connectés. Ce qui se traduit par des exigences assez importantes en termes de capacité. C'est dans ce cadre que les méthodes d'accès multiple non orthogonaux (NOMA) ont été introduit. Dans cette thèse, nous avons étudié et proposé une méthodes d'accès multiple basé sur la technique compute and forawrd et sur les réseaux de point (Lattice codes) tout en considérant différentes constructions de lattice. Nous avons également proposé des amélioration de l'algorithme de décodage de la méthode SCMA (Sparse code multiple access) basé sur les réseaux de points. Afin de simplifier les décodeurs multi-niveaux utilisés, nous avons proposé des expressions simplifiées de LLRs ainsi que des approximations. Finalement, nous avons étudié la construction D des lattices en utilisant les codes polaires. Cette thèse était en collaboration avec le centre de recherche de Huawei France. / Future generations of wireless networks pose many challenges for the research community. In particular, these networks must be able to respond, with a certain quality of service, to the demands of a large number of connected people and objects. This drives us into quite important requirements in terms of capacity. It is within this framework that non-orthogonal multiple access methods (NOMA) have been introduced. In this thesis, we have studied and proposed a multiple access method based on the compute and forward technique and on Lattice codes while considering different lattice constructions. We have also proposed improvements to the algorithm for decoding the Sparse code multiple access (SCMA) method based on Lattice codes. In order to simplify the multi-stage decoders used in here, we have proposed simplified expressions of LLRs as well as approximations. Finally, we studied the construction D of lattices using polar codes. This thesis was in collaboration with the research center of Huawei France.