Full 360 degree 3D finite element models are the most complete representation of Oil Country Tubular Goods (OCTG) premium threaded connections. Full 3D models can represent helical threads and boundary conditions required to simulate make-up and service loading. A methodology is developed to create a 360 degree full 3D parametric finite element model with helical threads as an effective design and analysis tool. The approach is demonstrated with the creation of a metal-to-metal seal integral joint full 3D model with manufacturer supplied geometry. The premium connection is decomposed into smaller parts to generate parametric geometric features. A controlled parametric meshing scheme is developed to manage mesh density in contact regions to effectively represent the mechanics in regions of interest while minimizing total element count. The scripted parametric approach allows for efficient geometric and mesh updates. Several methods to reduce and manage model runtimes are presented. An elastic-plastic material model is created with material coupon tensile tests results. Digital Image Correlation (DIC) is used to measure full-field displacement and strain data on the surface of the box. Experimental set up and data processing procedures are discussed. Error metrics are developed to correlate the finite element model results with the DIC experimental data. The DIC make-up experimental results are used to reconcile the finite element model to develop a minimum error make-up model relative to the pin rotation. The friction coefficient is estimated and the make-up torque-turn behavior is verified. The calibrated 3D finite element model is validated with ISO_13769 load series B axial and internal pressure loading experimental DIC data. Metal-to-metal seal metrics of contact pressure and seal length are evaluated. / Master of Science

La tendance actuelle des constructeurs pour le calcul scientifique est à l'utilisation de grappes de machines dont les noeuds comportent un nombre de coeurs toujours plus grand. Le modèle basé uniquement sur MPI laisse peu à peu la place à des modèles mélangeant l'utilisation de threads et de MPI. Ce changement de modèle entraîne de nombreuses problématiques car les implémentations MPI n'ont pas été conçues pour supporter les applications multi-threadées. Dans cette thèse, afin de garantir le bon fonctionnement des communications, nous proposons un module logiciel faisant interagir l'ordonnanceur de threads et la bibliothèque de communication. Ce gestionnaire d'entrées/sorties générique prend en charge la détection des événements du réseau et exploite les multiples unités de calcul présentes sur la machine de manière transparente. Grâce à la collaboration étroite avec l'ordonnanceur de threads, le gestionnaire d'entrées/sorties que nous proposons assure un haut niveau de réactivité aux événements du réseau. Nous montrons qu'il est ainsi possible de faire progresser les communications réseau en arrière-plan et donc de recouvrir les communications par du calcul. La parallélisation de la bibliothèque de communication est également facilité par un mécanisme d'exportation de tâches capable d'exploiter les différentes unités de calcul disponible tout en prenant en compte la localité des données. Les gains obtenus sur des tests synthétiques et sur des applications montre que l'interaction entre la bibliothèque de communication et l'ordonnanceur de threads permet de réduire le coût des communications et donc d'améliorer les performances d'une application. / The current trend of constructors for scientific computation is to build clusters whose node include an increasing number of cores.The classical programming model that is only based on MPI is being replaced by hybrid approaches that mix communication and multi-threading. This evolution of the programming model leads to numerous problems since MPI implementations were not designed for multi-threaded applications. In this thesis, in order to guarantee a smooth behavior of communication, we propose a software module that interact with both the threads scheduler and the communication library. This module, by working closely with the thread scheduler, allows to make communication progress in the background and guarantees a high level of reactivity to network events, even when the node is overloaded. We show that this permits to make communication progress in the background and thus to overlap communication and computation. The parallelization of the communication library is also made easier thanks to a task onloading mechanism that is able to exploit the available cores while taking data locality into account. The results we obtain on synthetic application as well as real-life applications show that the interaction between the thread scheduler and the communication library allows to reduce the overhead of communication and thus to improve the application performance.

To develop a high performant web service in Java, Reactive WebFlux has previously been utilized as the only alternative. With the introduction of virtual threads in Java 19, it has been considered that it would be an alternative to the current provided option. This paper presents the performance difference between Spring applications powered by normal threads, virtual threads, and its reactive counterpart Reactive WebFlux. The tests were done by creating three prototypes, which were used to call an endpoint with a set delay time and increasing the number of requests per second until system failure. The results show that the virtual threaded prototypes performed slightly better than the reactive prototype. The question remains whether the Reac- tive WebFlux application using the most optimal web server produces a different re- sult, as well as future research on how virtual threads perform with database usage. / För att utveckla högpresterande webbtjänster i Java har Reactive WebFlux varit det främsta alternativet. Med introduktionen av virtuella trådar i Java 19 har spekulerats kunna ersätta reaktiva webbtjänster. Detta arbete presenterar prestandaskillnaden mellan Spring-applikationer som drivs av normala trådar, virtuella trådar och dess reaktiva motsvarighet Reactive WebFlux. Testerna utfördes genom att skapa tre pro- totyper, som användes för att anropa en slutpunkt med en satt fördröjningstid och öka antalet förfrågningar per sekund fram till systemfel. Resultaten visade att proto- typen med virtuella trådar presterade något bättre än den reaktiva prototypen. Där- emot är det ännu oklart om Reactive WebFlux-applikationen med en optimal webb- server skulle producera ett annat resultat. Dessutom skulle det vara intressant för framtida forskning om hur virtuella trådar fungerar med databasanvändning.

A programação paralela tem sido freqüentemente adotada para o desenvolvimento de aplicações que demandam alto desempenho computacional. Com o advento das arquiteturas multi-cores e a existência de diversos níveis de paralelismo é importante definir estratégias de programação paralela que tirem proveito desse poder de processamento nessas arquiteturas. Neste contexto, este trabalho busca avaliar o desempenho da utilização das arquiteturas multi-cores, principalmente o oferecido pelas unidades de processamento gráfico (GPUs) e CPUs multi-cores na resolução de um problema térmico inverso. Algoritmos paralelos para a GPU e CPU foram desenvolvidos utilizando respectivamente as ferramentas de programação em arquiteturas de memória compartilhada NVIDIA CUDA (Compute Unified Device Architecture) e a API POSIX Threads. O algoritmo do método do gradiente conjugado pré-condicionado para resolução de sistemas lineares esparsos foi implementado totalmente no espaço da memória global da GPU em CUDA. O algoritmo desenvolvido foi avaliado em dois modelos de GPU, os quais se mostraram mais eficientes, apresentando um speedup de quatro vezes que a versão serial do algoritmo. A aplicação paralela em POSIX Threads foi avaliada em diferentes CPUs multi-cores com distintas microarquiteturas. Buscando um maior desempenho do código paralelizado foram utilizados flags de otimização as quais se mostraram muito eficientes na aplicação desenvolvida. Desta forma o código paralelizado com o auxílio das flags de otimização chegou a apresentar tempos de processamento cerca de doze vezes mais rápido que a versão serial no mesmo processador sem nenhum tipo de otimização. Assim tanto a abordagem utilizando a GPU como um co-processador genérico a CPU como a aplicação paralela empregando as CPUs multi-cores mostraram-se ferramentas eficientes para a resolução do problema térmico inverso. / Parallel programming has been frequently adopted for the development of applications that demand high-performance computing. With the advent of multi-cores architectures and the existence of several levels of parallelism are important to define programming strategies that take advantage of parallel processing power in these architectures. In this context, this study aims to evaluate the performance of architectures using multi-cores, mainly those offered by the graphics processing units (GPUs) and CPU multi-cores in the resolution of an inverse thermal problem. Parallel algorithms for the GPU and CPU were developed respectively, using the programming tools in shared memory architectures, NVIDIA CUDA (Compute Unified Device Architecture) and the POSIX Threads API. The algorithm of the preconditioned conjugate gradient method for solving sparse linear systems entirely within the global memory of the GPU was implemented by CUDA. It evaluated the two models of GPU, which proved more efficient by having a speedup was four times faster than the serial version of the algorithm. The parallel application in POSIX Threads was evaluated in different multi-core CPU with different microarchitectures. Optimization flags were used to achieve a higher performance of the parallelized code. As those were efficient in the developed application, the parallelized code presented processing times about twelve times faster than the serial version on the same processor without any optimization. Thus both the approach using GPU as a coprocessor to the CPU as a generic parallel application using the multi-core CPU proved to be more efficient tools for solving the inverse thermal problem.

[pt] O objetivo deste trabalho é explorar e estender a exibilidade provida pelo modelo híbrido de concorrência orientado a estágios, que visa integrar tanto loops de eventos cooperativos como threads preemptivas em um conceito único de mais alto nível. As contribuições deste trabalho estão centradas numa proposta de extensão do modelo de estágios que desacopla a especificação de aplicações concorrentes das decisões relacionadas ao ambiente de execução, permitindo que elas sejam fexivelmente mapeadas em diferentes configurações de acordo com as necessidades de escalonamento de tarefas e granularidade de processamento em partes específicas da aplicação. Procurando prover uma definição adequada para o conceito de modelos de concorrê- ncia híbridos, propomos um sistema de classificação que se baseia na combinação de características básicas de sistemas concorrentes e a possibilidade de execução paralela em múltiplos processadores. Com base nessa classificação, analisamos os benefícios e desvantagens associados a cada modelo de concorrência, justificando a adoção de modelos que combinam threads e eventos em um mesmo ambiente de programação, e descrevemos a extensão do paradigma de programação orientado a estágios. Finalmente, apresentamos a implementação do modelo proposto na linguagem Lua e seu uso em cenários de execução que confirmam os benefícios da extensão do modelo de estágios na especificação de aplicações concorrentes. / [en] The purpose of this work is to explore and extend the flexibility provided by the staged event-driven concurrency model to integrate both cooperative event loops and preemptive threads in a single high level abstraction. The contributions of this work are focused on the extension of the staged model that decouples the specification of concurrent applications of the decisions related to the execution environment, allowing them to be fl exibly mapped to different configurations according to the task scheduling needs and processing granularity in specific parts of the application. In order to provide an adequate definition of the concept of hybrid concurrency models, we propose a classification system that is based on the combination of basic features of concurrent systems and the possibility of parallel execution on multiple processors. Based on this classification, we analyze the benefits and drawbacks associated with each concurrency model, justifying the adoption of models that combine threads and events in the same programming environment and the extension of the staged model. Finally, we present the implementation of the proposed model in the Lua programming language and its use in execution scenarios that confirm the benefits of the extension of the staged model in the specification of concurrent applications.

[pt] Controle de sobrecarga pode ser feito com o uso de políticas de escalonamento adequadas, que procuram ajustar dinamicamente os recursos alocados a uma aplicação. Pela dificuldade de implementação, muitas vezes desenvolvedores se veem obrigados a reprogramar o sistema para adequá-lo a uma determinada política. Através do estudo de diversas políticas de escalonamento, propomos neste trabalho um modelo de interface que permite a criação e monitoração de novas políticas dentro de arquiteturas baseadas em estágios. Implementamos a interface de programação proposta e exercitamos um conjunto de políticas que construímos sobre ela em duas aplicações com características de carga bem distintas. / [en] Specific scheduling policies can be appropriate for overload control in different application scenarios. However, these policies are often difficult to implement, leading developers to reprogram entire systems in order to adapt them to a particular policy. Through the study of various scheduling policies, we propose an interface model that allows the programmer to integrate new policies and monitoring schemes to the same application in a Staged Event-Driven Architecture. We describe the implementation of the proposed interface and the results of it s use in implementing a set of scheduling policies for two applications with different load profiles.

A programação paralela tem sido freqüentemente adotada para o desenvolvimento de aplicações que demandam alto desempenho computacional. Com o advento das arquiteturas multi-cores e a existência de diversos níveis de paralelismo é importante definir estratégias de programação paralela que tirem proveito desse poder de processamento nessas arquiteturas. Neste contexto, este trabalho busca avaliar o desempenho da utilização das arquiteturas multi-cores, principalmente o oferecido pelas unidades de processamento gráfico (GPUs) e CPUs multi-cores na resolução de um problema térmico inverso. Algoritmos paralelos para a GPU e CPU foram desenvolvidos utilizando respectivamente as ferramentas de programação em arquiteturas de memória compartilhada NVIDIA CUDA (Compute Unified Device Architecture) e a API POSIX Threads. O algoritmo do método do gradiente conjugado pré-condicionado para resolução de sistemas lineares esparsos foi implementado totalmente no espaço da memória global da GPU em CUDA. O algoritmo desenvolvido foi avaliado em dois modelos de GPU, os quais se mostraram mais eficientes, apresentando um speedup de quatro vezes que a versão serial do algoritmo. A aplicação paralela em POSIX Threads foi avaliada em diferentes CPUs multi-cores com distintas microarquiteturas. Buscando um maior desempenho do código paralelizado foram utilizados flags de otimização as quais se mostraram muito eficientes na aplicação desenvolvida. Desta forma o código paralelizado com o auxílio das flags de otimização chegou a apresentar tempos de processamento cerca de doze vezes mais rápido que a versão serial no mesmo processador sem nenhum tipo de otimização. Assim tanto a abordagem utilizando a GPU como um co-processador genérico a CPU como a aplicação paralela empregando as CPUs multi-cores mostraram-se ferramentas eficientes para a resolução do problema térmico inverso. / Parallel programming has been frequently adopted for the development of applications that demand high-performance computing. With the advent of multi-cores architectures and the existence of several levels of parallelism are important to define programming strategies that take advantage of parallel processing power in these architectures. In this context, this study aims to evaluate the performance of architectures using multi-cores, mainly those offered by the graphics processing units (GPUs) and CPU multi-cores in the resolution of an inverse thermal problem. Parallel algorithms for the GPU and CPU were developed respectively, using the programming tools in shared memory architectures, NVIDIA CUDA (Compute Unified Device Architecture) and the POSIX Threads API. The algorithm of the preconditioned conjugate gradient method for solving sparse linear systems entirely within the global memory of the GPU was implemented by CUDA. It evaluated the two models of GPU, which proved more efficient by having a speedup was four times faster than the serial version of the algorithm. The parallel application in POSIX Threads was evaluated in different multi-core CPU with different microarchitectures. Optimization flags were used to achieve a higher performance of the parallelized code. As those were efficient in the developed application, the parallelized code presented processing times about twelve times faster than the serial version on the same processor without any optimization. Thus both the approach using GPU as a coprocessor to the CPU as a generic parallel application using the multi-core CPU proved to be more efficient tools for solving the inverse thermal problem.

This thesis investigates the effect of incorporating various information visualisation techniques and non-speech sounds (i.e. auditory icons and earcons) in email browsing. This empirical work consisted of three experimental phases. The first experimental phase aimed at finding out the most usable visualisation techniques for presenting email information. This experiment involved the development of two experimental email visualisation approaches which were called LinearVis and MatrixVis. These approaches visualised email messages based on a dateline together with various types of email information such as the time and the senders. The findings of this experiment were used as a basis for the development of a further email visualisation approach which was called LinearVis II. This novel approach presented email data based on multi-coordinated views. The usability of messages retrieval in this approach was investigated and compared to a typical email client in the second experimental phase. Users were required to retrieve email messages in the two experiments with the provided relevant information such as the subject, status and priority. The third experimental phase aimed at exploring the usability of retrieving email messages by using other type of email data, particularly email threads. This experiment investigated the synergic use of graphical representations with non-speech sounds (Multimodal Metaphors), graphical representations and textual display to present email threads and to communicate contextual information about email threads. The findings of this empirical study demonstrated that there is a high potential for using information visualisation techniques and non-speech sounds (i.e. auditory icons and earcons) to improve the usability of email message retrieval. Furthermore, the thesis concludes with a set of empirically derived guidelines for the use of information visualisation techniques and non-speech sound to improve email browsing.

This thesis provides a tour through an imaginary building that contains the work I have completed in the last two years at Virginia Commonwealth University. The body of the text provides a discussion of specific paintings as well as more general themes related to painting and art. The discussion includes thoughts on futility, desire, schism, the leap, collage, photography, materiality, painting, image, and landscape. The second part of the text is an abstract statement about the paintings included in the thesis show, Splinter Paintings.

Threaded Behavioral Protocols (TBP) is a specification language for modelling the behavior of software components. This thesis aims at an analysis of TBP specifications within environments which involve an unbounded replication of threads. Such a TBP specification - together with a model of the possible environments - induces infinite state space which contains a vast amount of symmetries caused by thread replication. A model checking technique addressing such a state space and reducing the symmetries by using symbolic counter abstraction is proposed. In order to utilize the symbolic counter abstraction, the properties of the TBP specifications (called provisions) are converted into thread state reachability properties. The proposed analysis is safe in the sense that it discovers all errors in the model. On the other hand, it may yield spurious errors, i.e., errors that do not correspond to any real error in the model. The spurious errors are well identified and further possibilities to reduce them are outlined. Beyond the scope of the specific specifications, this work may also present a small step towards supporting dynamic thread creation in TBP.