A recognition system is required between pinniped mothers and pups. For otariids this is especially important since females frequently leave their pups for foraging and must reunite on return. Pups must deal with these fasting periods during maternal absence and consequently may attempt to obtain allomaternal care from unrelated females. This research on the New Zealand fur seal (Arctocephalus forsteri) at Ohau Point, Kaikoura, New Zealand, quantified mother-pup recognition behaviour during reunions, individuality of pup calls used by mothers to recognise their pup, and the occurrence of allosuckling as a possible recognition error by females and as a strategy employed by pups to gain allomaternal care during their mothers' absence. A combination of behavioural observations, morphometry, VHF radio telemetry, acoustics and DNA genotyping were employed to study these topics. Postpartum interaction behaviours between mothers and pups appeared to facilitate development of an efficient mother-pup recognition system, involving mainly vocal and olfactory cues that were utilised during reunions. Greater selective pressure on pups to reunite resulted in an asymmetry of searching behaviour between females and pups during reunions. The vocalisations of pups were stereotypic, especially those features of the fundamental frequency and frequency of the lowest harmonic, which are likely to facilitate recognition of a pup by their mother. Pups attempted to steal milk from unrelated females more often during maternal absence and appeared to modify the intra-individual variation pattern of a feature of their vocal signatures over this period, which may assist attempts at allosuckling under nutritional stress. Fostering was demonstrated to occur despite costs to filial pups and possible costs to female reproductive success and may be attributed to development of erroneous recognition between females and non filial pups, or kin selection. This study provides a valuable contribution to the knowledge of recognition systems between pinniped mothers and pups, of alternative pup strategies under nutritional stress and of the rare occurrence of fostering in otariid pinnipeds.

Com o objetivo de ser portável e eficiente em arquiteturas HPC atuais, a execução de um programa paralelo deve ser adaptável. Este trabalho mostra como isso pode ser atingido utilizando MPI, através de criação dinâmica de processos, integrada com programação Divisão-e-Conquista e uma estratégia Work-Stealing para balancear os processos MPI, em ambientes heterogêneos e/ou dinâmicos, em tempo de execução. Este trabalho explica como implementar uma aplicação segundo o modelo de Divisão-e-Conquista com MPI, bem como a implementação de uma estratégia Work-Stealing. São apresentados resultados experimentais baseados em uma aplicação sintética, o problema das N-Rainhas (N-Queens). Valida-se tanto a adaptabilidade e a eficiência do código. Os resultados mostram que é possível utilizar um padrão amplamente difundido como o MPI, mesmo em plataformas de HPC não tão homogêneas como um cluster. / In order to be portable and efficient on modern HPC architectures, the execution of a parallel program must be adaptable. This work shows how to achieve this in MPI, by the dynamic creation of processes, coupled with Divide-and-Conquer programming and a Work-Stealing strategy to balance the MPI processes, in a heterogeneous and/or dynamic environment, at runtime. The application of Divide and Conquer with MPI is explained, as well as the implementation of a Work-Stealing strategy. Experimental results are provided, based on a synthetic application, the N-Queens computation. Both the adaptability of the code and its efficiency are validated. The results show that it is possible to use widely spread standards such as MPI, even in parallel HPC platforms that are not as homogeneous as a Cluster.

Com o objetivo de ser portável e eficiente em arquiteturas HPC atuais, a execução de um programa paralelo deve ser adaptável. Este trabalho mostra como isso pode ser atingido utilizando MPI, através de criação dinâmica de processos, integrada com programação Divisão-e-Conquista e uma estratégia Work-Stealing para balancear os processos MPI, em ambientes heterogêneos e/ou dinâmicos, em tempo de execução. Este trabalho explica como implementar uma aplicação segundo o modelo de Divisão-e-Conquista com MPI, bem como a implementação de uma estratégia Work-Stealing. São apresentados resultados experimentais baseados em uma aplicação sintética, o problema das N-Rainhas (N-Queens). Valida-se tanto a adaptabilidade e a eficiência do código. Os resultados mostram que é possível utilizar um padrão amplamente difundido como o MPI, mesmo em plataformas de HPC não tão homogêneas como um cluster. / In order to be portable and efficient on modern HPC architectures, the execution of a parallel program must be adaptable. This work shows how to achieve this in MPI, by the dynamic creation of processes, coupled with Divide-and-Conquer programming and a Work-Stealing strategy to balance the MPI processes, in a heterogeneous and/or dynamic environment, at runtime. The application of Divide and Conquer with MPI is explained, as well as the implementation of a Work-Stealing strategy. Experimental results are provided, based on a synthetic application, the N-Queens computation. Both the adaptability of the code and its efficiency are validated. The results show that it is possible to use widely spread standards such as MPI, even in parallel HPC platforms that are not as homogeneous as a Cluster.

Com o objetivo de ser portável e eficiente em arquiteturas HPC atuais, a execução de um programa paralelo deve ser adaptável. Este trabalho mostra como isso pode ser atingido utilizando MPI, através de criação dinâmica de processos, integrada com programação Divisão-e-Conquista e uma estratégia Work-Stealing para balancear os processos MPI, em ambientes heterogêneos e/ou dinâmicos, em tempo de execução. Este trabalho explica como implementar uma aplicação segundo o modelo de Divisão-e-Conquista com MPI, bem como a implementação de uma estratégia Work-Stealing. São apresentados resultados experimentais baseados em uma aplicação sintética, o problema das N-Rainhas (N-Queens). Valida-se tanto a adaptabilidade e a eficiência do código. Os resultados mostram que é possível utilizar um padrão amplamente difundido como o MPI, mesmo em plataformas de HPC não tão homogêneas como um cluster. / In order to be portable and efficient on modern HPC architectures, the execution of a parallel program must be adaptable. This work shows how to achieve this in MPI, by the dynamic creation of processes, coupled with Divide-and-Conquer programming and a Work-Stealing strategy to balance the MPI processes, in a heterogeneous and/or dynamic environment, at runtime. The application of Divide and Conquer with MPI is explained, as well as the implementation of a Work-Stealing strategy. Experimental results are provided, based on a synthetic application, the N-Queens computation. Both the adaptability of the code and its efficiency are validated. The results show that it is possible to use widely spread standards such as MPI, even in parallel HPC platforms that are not as homogeneous as a Cluster.

A combination of an original work of fiction, Alone Among Friends, and a critical discussion of masculinity in the work of Per Petterson, this dissertation joins a growing conversation in the field of Masculine Studies about the depiction of men in literature. Written in a spare and realistic style, Alone Among Friends is a novella that hopes to explore ideas of masculinity, friendship, success, and failure present in the mindset of the American Millennial generation. Takings its cues from The Sun Also Rises, Light Years, and The Salt Point, Alone Among Friends examines the destructive nature of hyper-masculinity and highlights the danger of attaching too much meaning to external validation as the measuring stick for one's self worth. Moreover, Alone Among Friends is also influenced by the themes of memory and knowing found within the work of Per Petterson. "How Memories of the Father Inform a Son's Understanding of Masculinity in the Novels of Per Petterson" discusses the ways in which Per Petterson, a Norwegian writer, has been both influenced by American notions of masculinity and also managed to incorporate European aspects of family into his work to create a unique hybrid perspective that merges the American idea of the emancipated male protagonist with the European family centered narrative. By tracing Petterson's influences, this dissertation will attempt to show how Petterson is a logical heir to such American writers as Ernest Hemingway and Richard Ford, and it will argue that, by firmly grounding his novels in the point of view of sons that study every aspect of their fathers, Petterson inhabits a space that has long been absent from both American and Norwegian literature alike. Through the examination of Out Stealing Horses and In the Wake, this dissertation focuses on the varying ways in which these direct comparisons between father and son influence the son's perception of his own success and failure as a man.

Nous étudions la parallélisation de la procédure de recherche de solution d’un problème en Programmation Par Contraintes (PPC). Après une étude de l’état de l’art, nous présentons une nouvelle méthode, nommée Embarrassingly Parallel Search (EPS). Cette méthode est basée sur la décomposition d’un problème en un très grand nombre de sous-problèmes disjoints qui sont ensuite résolus en parallèle par des unités de calcul avec très peu, voire aucune communication. Le principe d’EPS est d’arriver statistiquement à un équilibrage des temps de résolution de chaque unité de calcul afin d’obtenir une bonne répartition de la charge de travail. EPS s’appuie sur la propriété suivante : la somme des temps de résolution de chacun des sous-problèmes est comparable au temps de résolution du problème en entier. Cette propriété est vérifiée en PPC, ce qui nous permet de disposer d’une méthode simple et efficace en pratique. Dans nos expérimentations, nous nous intéressons à la recherche de toutes les solutions d’un problème en PPC, à prouver qu’un problème n’a pas de solution et à la recherche d’une solution optimale d’un problème d’optimisation. Les résultats montrent que la décomposition doit générer au moins 30 sous-problèmes par unité de calcul pour obtenir des charges de travail par unité de calcul équivalentes. Nous évaluons notre approche sur différentes architectures (machine multi-coeurs, centre de calcul et cloud computing) et montrons qu’elle obtient un gain pratiquement linéaire en fonction du nombre d’unités de calcul. Une comparaison avec les méthodes actuelles telles que le work stealing ou le portfolio montre qu’EPS obtient de meilleurs résultats. / We study the search procedure parallelization in Constraint Programming (CP). After giving an overview on various existing methods of the state-of-the-art, we present a new method, named Embarrassinqly Parallel Search (EPS). This method is based on the decomposition of a problem into many disjoint subproblems which are then solved in parallel by computing units with little or without communication. The principle of EPS is to have a resolution times balancing for each computing unit in a statistical sense to obtain a goodDépôt de thèse – Données complémentaireswell-balanced workload. We assume that the amount of resolution times of all subproblems is comparable to the resolution time of the entire problem. This property is checked with CP and allows us to have a simple and efficient method in practice. In our experiments, we are interested in enumerating all solutions of a problem, and proving that a problem has no solution and finding an optimal solution of an optimization problem. We observe that the decomposition has to generate at least 30 subproblems per computing unit to get equivalent workloads per computing unit. Then, we evaluate our approach on different architectures (multicore machine, cluster and cloud computing) and we observe a substantially linear speedup. A comparison with current methods such as work stealing or portfolio shows that EPS gets better results.

Barrelfish operating system is an experimental instance of multikernel structure which exhibits good features such as hardware heterogeneity, scalability, dynamicity, etc. Barrelfish is in progress and lacks applications. Therefore, there is a need to investigate the efficiency of applications running in Barrelfish and one of candidates is a shared-memory application. To conduct an empirical study, Cilk is chosen inasmuch as its runtime library is designed for shared-memory architectures and it has been known to expose good performance. This thesis focuses on making Cilk run on top of Barrelfish in order to reach two goals: portability which is described to be supported by Barrelfish, and good speed afterwards. The porting involves compiling Cilk runtime source code by replacing its pthread subroutines with set of APIs in Barrelfish and then changing the way Cilk scheduler spawns worker thread on multiple cores. However, the main point of the porting is to make different cores access to the same virtual address space. Luckily, Barrelfish provides a notion of domain which specifies the number of cores in an application so that these cores can share the same memory space. This thesis also has carried out benchmarks on some Cilk programs and found that Cilk does not perform as well as it is expected. In addition measurements on parallel workers shows that Cilk on Barrelfish takes more cycles to perform computation. Although Cilk still maintains work-first principle, it cannot achieve the time bound. The spanning domain cost is proportional to the number of cores, but it will matter if applications take small time to complete.

Un Sistema de Tiempo Real (STR) debe garantizar que sus resultados, además de correctos, cumplan también con un conjunto de restricciones temporales. En general, esto implica asegurar que sus tareas finalicen su ejecución antes de un vencimiento. Para cumplir esto, la predictibilidad y el determinismo adquieren suma importancia. El campo de aplicación clásico de los STR ha sido la industria, como por ejemplo la aviónica, la exploración espacial, equipamiento médico, sistemas de control, etc. Todos estos sistemas tienen en común el de ser de misión crítica, donde un fallo tiene consecuencias graves, como pérdidas materiales y económicas, daños al medio ambiente o poner en riesgo la vida humana. Por lo general estos sistemas son estáticos, y utilizan arquitecturas de hardware y algoritmos de efectividad comprobada. En muchas ocasiones su diseño e implementación es ad-hoc. Sin embargo, en las últimas décadas el campo de aplicación de los STR se ha extendiendo. En la actualidad son utilizados en ámbitos y productos de lo más variados: electrodomésticos, productos electrónicos de consumo, telefonía celular, automóviles, comunicaciones, sistemas de reservas de pasajes, etc. Muchos de estos sistemas están constituidos tanto por tareas críticas como por tareas no-críticas. Un fallo en la ejecución de las primeras tiene consecuencias graves, en tanto que el incumplimiento de las restricciones temporales de las últimas afecta la calidad de servicio esperada. Es vital entonces que las tareas no-críticas no afecten la correcta ejecución de las tareas críticas. Un STR con estas características se denomina heterogéneo. En los últimos años, gracias al incremento de la potencia de cálculo de los microprocesadores, y la reducción de sus costos, el número de STR que permiten que coexistan estos dos conjuntos de tareas ha aumentado. Para lograr una correcta ejecución de un STR heterogéneo, se requiere de técnicas que calculen y administren en línea, el tiempo ocioso disponible. De esta manera, el planificador puede mantener la garantía decumplimiento de las constricciones temporales de las tareas críticas, y al mismo tiempo brindar una atención aceptable a las tareas sin requerimientos estrictos. En la actualidad, microprocesadores más potentes abren la posibilidad de implementar estos métodos incluso en sistemas que antaño contaban con muy baja potencia de cálculo. Sin embargo, la sobrecarga que agregan no es despreciable, por lo que reducir el costo computacional de estos métodos sigue siendo de suma utilidad, aún cuando se dispone de hardware con mayor capacidad de computo. Existe una amplia literatura que aborda la problemática de la planificación de STR heterogéneos. Sin embargo, existe una brecha significativa entre los desarrollos teóricos en la disciplina, y las técnicas efectivamente utilizadas en la industria. Es necesario poder comprobar el costo real y las ventajas y desventajas objetivas de implementar los modelos teóricos de punta. Muchos modelos teóricos no tienen en cuenta costos adicionales presentes en implementaciones concretas. Estos son comúnmente considerados despreciables en la modelización, a fin de simplificar el análisis, la evaluación y el desarrollo del sistema. Como consecuencia, en la implementación real se estos parámetros se sobrestiman, lo que resulta en una menor eficiencia del sistema. Un ejemplo común es el uso de microprocesadores con una capacidad de cálculo por encima de la realmente requerida, lo que impacta negativamente en el consumo de energía y en los costos. Un modelo más detallado en la etapa de diseño, implementación y verificación, permitiría mejorar el desempeño del sistema final, sin abandonar la garantía de predictibilidad temporal. Igualmente importantes, se deben contar con técnicas y herramientas que permitan implementar estos modelos métodos teóricos de manera eficiente. La presente tesis se basa en la hipótesis de que los STR heterogéneos pueden ser eficaces en la planificación de sus conjuntos de tareas y en el uso de sus recursos computacionales. Bajo esta premisa, se presentan nuevos modelos y mejoras a modelos ya existentes, junto con simulaciones, pruebas y desarrollos necesarios para su verificación. El trabajo se basa fuertemente en la implementación práctica de los resultados teóricos, identificando las dificultades reales que su puesta en práctica trae aparejado. De esta manera, se busca reducir la brecha existente entre los resultados obtenidos por la investigación científica en la disciplina de los STR, y aquello concretamente utilizado e implementado realmente en la industria, la investigación y el desarrollo tecnológico. / A Real-Time System (RTS) must warrant that its results are correct and fulfill a set of temporal restrictions. This implies that each task completes its execution before a specific deadline. In order to accomplish this, the predictibility and determinism of the system as a whole is very important. These kind of systems are used in several industries, like aircraft avionics, space exploration, medical equipment, etc., which are mission critical. A failure in this systems could have catastrophic consequences, like loss of human lives. Most of the time the design and implementation of these systems is ad-hoc. In the last decades, thanks to the growth and sophistication of embedded systems, the application domain of the RTS increased. Nowdays they can be found on consumer electronics, cellphones, communications systems, cars, etc. A lot of these new kind of real-time systems are composed of both critical and noncritical tasks. A failure in the execution of the former have severe consequences, but a missed deadline of the later only affects the quality of service. Such a RTS is known as a heterogeneus one. In order to accomplish a correct execution of a heterogeneus RTS, methods and techniques that calculates and manages the system idle-time are needed. With these tools, the system scheduler can guarantee that all the time-critical tasks fulfill their deadlines. Nonetheless, these techniques add an execution overhead to the system. Although severalworks in the literature proposes solutions for many of the scheduling problems in a heterogeneus RTS, a gap exists between these results and what is actually used and implemented in the industry. Many theoretical models do not take into account the additional costs present in a concrete implementation. These are commonly considered negligible in order to simplify the analysis, evaluation and development of the system. As a consequence, some parameters are overestimated, resulting in reduced system efficiency. A common scenario is the use of microprocessors more powerful than required, with negative impact on energy consumption and production costs. A more detailed model in the design and implementation stage could improve the performance of the final system, without abandoning the guarantee of temporal predictability. Equally important, there must be techniques and tools that allow the implementation of these theoretical results. The working hipothesis of this thesis is that a heterogeneus RTS could be efficient in the scheduling of their tasks and resources. Following this premise, new models and improvements to existing ones are presented, in conjunction with several simulations and implementations of the theoretical results, in order to identify the real difficulties that the implementation brings about. This seeks to reduce the gap between the scientific research in the discipline of RTS and what actually implemented in the industry.

Nos últimos anos, uma das alternativas adotadas para aumentar o desempenho de sistemas de processamento de alto desempenho têm sido o uso de arquiteturas híbridas. Essas arquiteturas são constituídas de processadores multicore e coprocessadores especializados, como GPUs. Esses coprocessadores atuam como aceleradores em alguns tipos de operações. Por outro lado, as ferramentas e modelos de programação paralela atuais não são adequados para cenários híbridos, produzindo aplicações pouco portáveis. O paralelismo de tarefas considerado um paradigma de programação genérico e de alto nível pode ser adotado neste cenário. Porém, exige o uso de algoritmos de escalonamento dinâmicos, como o algoritmo de roubo de tarefas. Neste contexto, este trabalho apresenta um middleware (WORMS) que oferece suporte ao paralelismo de tarefas com escalonamento por roubo de tarefas em sistemas híbridos multi-CPU e multi-GPU. Esse middleware permite que as tarefas tenham implementação tanto para execução em CPUs quanto em GPUs, decidindo em tempo de execução qual das implementações será executada de acordo com os recursos de hardware disponíveis. Os resultados obtidos com o WORMS mostram ser possível superar, em algumas aplicações, tanto o desempenho de ferramentas de referência para execução em CPU quanto de ferramentas para execução em GPUs. / In the last years, one of alternatives adopted to increase performance in high performance computing systems have been the use of hybrid architectures. These architectures consist of multicore processors and specialized coprocessors, like GPUs. Coprocessors act as accelerators in some types of operations. On the other hand, current parallel programming models and tools are not suitable for hybrid scenarios, generating less portable applications. Task parallelism, considered a generic and high level programming paradigm, can be used in this scenario. However, it requires the use of dynamic scheduling algorithms, such as work stealing. In this context, this work presents a middleware (WORMS) that supports task parallelism with work stealing scheduling in multi-CPU and multi-GPU systems. This middleware allows task implementations for both CPU and GPU, deciding at runtime which implementation will run according to the available hardware resources. The performance results obtained with WORMS showed that is possible to outperform both CPU and GPU reference tools in some applications.

This Special Report attempts to capture, in a very readable journalistic style, the fundamental complexity of the conflict in the Weenen District. It is an honest attempt to look beyond the comfortable stereotypes. Hopefully, it will contribute to a clearer understanding of attitudes to land and the conflict around land. There are no easy solutions to this problem. But, hopefully, this report will assist in the process of finding workable solutions to land struggles in Weenen and other neglected parts of South Africa. / Special Report no. 8