O desempenho global dos sistemas computacionais é limitado, geralmente, pelo componente de menor desempenho. Os processadores e a memória principal têm experimentado um aumento de desempenho bem maior que o da memória secundária, como os discos magnéticos. Em 1984, Johnson introduziu o conceito de fragmentação, onde um dado é gravado em uma matriz de discos, de forma que os seus fragmentos podem ser recuperados em paralelo e, por conseqüência, de forma mais rápida. O principal problema da fragmentação é a redução da confiabilidade da matriz pois, a falha de um dos discos torna o dado inacessível. Patterson, Gibson e Katz propuseram, em 1988, 5 formas de armazenar informação redundante na matriz de discos e, dessa forma, aumentar sua confiabilidade. A essas formas foi dado o nome de RAID - Redundant Arrays of Independent Disks. Com o passar do tempo, outras formas de armazenamento de redundância foram criadas, tornando complexa a taxonomia da área. Além disso, alterações de parâmetros na matriz implicam em variações de desempenho nem sempre fáceis de se perceber em um primeiro momento. Com o objetivo de facilitar a compreensão da taxonomia e permitir que sejam feitos experimentos na matriz buscando um melhor desempenho, esta dissertação propõe um ambiente de simulação e aprendizado para RAID, onde o usuário pode interagir com diversos modelos de RAID, ou até criar o seu próprio, para avaliar seu desempenho em várias situações, além de oferecer ao usuário acesso ao conhecimento da área, agindo como um tutor. Esta dissertação apresenta, ainda, um protótipo de um simulador de discos magnéticos que pode ser utilizado como base para o desenvolvimento de um simulador de RAID para ser utilizado pelo ambiente. / The component with the worst performance usually limits the overall performance of a computing system. The performance of processors and main memory has improved faster than the secondary memory, such as magnetic disks. Johnson, in 1984, introduced the concept of fragmentation, in which a data file is written into a disk array, in a way that its stripes can be recovered in parallel and therefore, in a faster way. The main problem with fragmentation is the reduction of the reliability. If one disk fails, all data file becomes inaccessible. Patterson, Gibson and Katz proposed, in 1988, five ways to store redundant information in the array, increasing the reliability, comprising the main RAID (Redundant Array of Independent Disks) configurations. Some other ways to store the redundant information have been proposed over the years, making the RAID taxonomy more complex. Furthermore, changes in the array parameters takes to performance variations that are not always understood. With the purpose of facilitating the comprehension of the taxonomy and allowing the execution of experiments looking forward to improve performance, this MSc Dissertation proposes an Intelligent Simulation and Learning Environment for RAID, where the user can interact with several RAID models, or even create his/her own models, in order to evaluate their performance under different situations. The environment also allows the user to interact with the knowledge of the area, acting as a tutor. This Dissertation also presents a prototype of a magnetic disk simulator, that can be used as the kernel for the development of a RAID simulator to be used by the environment.

Thesis advisor: Jan R. Engelbrecht / Thesis advisor: Renato E. Mirollo / Many phenomena in nature that involve ordering in time can be understood as collective behavior of coupled oscillators. One paradigm for studying a population of self-sustained oscillators is the Kuramoto model, where each oscillator is described by a phase variable, and interacts with other oscillators through trigonometric functions of phase differences. This dissertation studies $N$ identical Kuramoto oscillators in a general form \[ \dot{\theta}_{j}=A+B\cos\theta_{j}+C\sin\theta_{j}\qquad j=1,\dots,N, \] where coefficients $A$, $B$, and $C$ are symmetric functions of all oscillators $(\theta_{1},\dots,\theta_{N})$. Dynamics of this model live in group orbits of M\"obius transformations, which are low-dimensional manifolds in the full state space. When the system is a phase model (invariant under a global phase shift), trajectories in a group orbit can be identified as flows in the unit disk with an intrinsic hyperbolic metric. A simple criterion for such system to be a gradient flow is found, which leads to new classes of models that can be described by potential or Hamiltonian functions while exhibiting a large number of constants of motions. A generalization to extended phase models with non-identical couplings gives rise to richer structures of fixed points and bifurcations. When the coupling weights sum to zero, the system is simultaneously gradient and Hamiltonian. The flows mimic field lines of a two-dimensional electrostatic system consisting of equal amounts of positive and negative charges. Bifurcations on a partially synchronized subspace are discussed as well. / Thesis (PhD) — Boston College, 2017. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Physics.

Este trabalho apresenta um estudo sobre o gargalo no acesso a disco sob arquiteturas com sistemas multiprocessadores e processadores multicore e apresenta uma proposta de solução para o problema e sua implementação. A solução desenvolvida é chamada de FPA. A estratégia desenvolvida no trabalho aproveita os recursos de memória compartilhada da arquitetura e utiliza buffers de dados compartilhados entre as diversas threads das aplicações, bem como tenta otimizar o acesso ao disco rígido. Para que seja diminuída a concorrência no acesso a disco, é permitido que apenas uma thread realize leituras ou gravações no disco por vez, e nas situações de escrita é feita a sobreposição entre o processamento e o envio dos dados ao disco. As operações de escrita são realizadas por uma thread adicional, que é inicializada e gerenciada pela FPA. Foram executados diversos testes com diferentes números de threads, tamanhos de arquivo e quantidade de processamento de cálculos na aplicação. Os resultados apresentaram ganho de desempenho com a FPA em diversas situações, onde um dos fatores mais significativos para o ganho de desempenho é a quantidade de processamento existente na aplicação. De modo geral, as operações de leitura apresentaram ganho de desempenho em relação ao uso de streams nas situações onde existe pouco processamento de cálculos na aplicação, e as operações de escrita mostraram vantagem nas situações onde existe uma quantidade de processamento de cálculos maior na aplicação. Nas operações de escrita é obtido vantagem com a sobreposição entre processamento e escrita de dados. Porém, quando a quantidade de processamento de cálculos é muito alta, o tempo de espera pela gravação em disco torna-se irrisório perante o tempo gasto no processamento dos cálculos, ocasionando perda de desempenho da FPA em relação aos streams. / This work presents a study on disk access bottlenecks for multiprocessor architecture systems and also proposes a specification to solve this problem and its implementation. The developed strategy is called FPA. The developed strategy makes use of shared memory architecture and uses shared data buffers among several application threads and also tries to optimize disk access time. In order to decrease disk concurrency, is allowed to just one application thread to read/write simultaneously and processing thread operations run parallelly to disk operations. Write operations are performed by an internal thread, which is started and managed by FPA. Several tests were performed concerning different amounts of application threads instances, file sizes and computing instructions. Results show performance gain with FPA in several situations. The main factor that affects performance is the amount of computing instructions on each application. In general, read operations showed a better performance compared to use of streams in situations where there is few computing instructions in the application, and write operations showed advantages on situations where there are more computing instructions in the application. Write operations are advantageous using data writing and computing instructions overlap. However, in cases of a large amount of computing instructions compared to disk operations, the time to perform disk operations becomes too few compared to computing time, occurring FPA performance loss compared to streams.

Among the numerous taphonomic influences that can impact biological remains, agricultural activity has one of the most widespread effects. The present research examines the impacts of agricultural activities on buried skeletal remains, both in archaeological and medicolegal contexts. Juvenile pig (Sus scrofa) skeletons were utilized to simulate buried juvenile human aged 3 to 5 years to test the influence of original burial depths and different plowing intervals on the dispersal and the degree of damage caused by an offset disk plow. Ten juvenile pig skeletons were buried in relatively anatomical position, five at a bottom depth of 15 cmbs (cm below the surface) and five at 22 cmbs. They were subjected to different intervals of plowing with one burial at each depth subjected to a single, three, five, seven, or ten plow passes. The disturbed area was surveyed for surface material, and the plow furrows were excavated in 1 m by 1 m units. The excavated soil was screened using a ¼" mesh screen, except for two burials that were screened using a nested ¼" and ⅛" mesh screen to test the differences in recovery between the two screen sizes. The recovered skeletal material was inventoried and assessed for damage. The results of this research showed that while original burial depth had some effect on distribution of bone and the degree to which bone was damaged, the relationship was not statistically significant. The number of plowing intervals did have a significant effect on the distribution and degree of damage. The damage caused to bone by the offset disk plow would be distinguishable from perimortem trauma in dry bone. Comparison of the recovery rates of ¼" and ⅛" mesh found that ⅛" contributed to a higher recovery of juvenile remains at least 3 to 5 years of age in soil with a high gravel content.

O elemento químico neodímio (Nd) atualmente está em evidência pois é parte fundamental em aplicações produzidas por empresas ligadas a tecnologias limpas e de ponta. Porém, é classificado como crítico pelo departamento de energia dos EUA em função do monopólio chinês atuante em sua exploração e comercialização. Devido à crescente demanda frente à restrita disponibilidade, a recuperação deste metal está em ascensão. Uma das aplicações do neodímio é o ímã permanente de neodímio-ferro-boro, utilizado em hard disk drives (HDs) de computadores. Desta forma, este estudo visou caracterizar os ímãs provenientes de HDs de computadores e estabelecer uma rota de recuperação do neodímio por meio de processos mecânicos e hidrometalúrgicos. Por meio da pesagem dos HDs e seus ímãs obteve-se massa média de 3,04 g e 7,91 g para laptop e desktop, respectivamente, representando menos de 3% da massa total do HD. A desmagnetização foi atingida por meio de aquecimento até a temperatura de Curie, em torno de 312 °C, e a cominuição realizada em moinho de facas, onde foram obtidas três granulometrias classificadas em frações A, B e C. Qualitativamente as amostras, foram analisados por meio de difração de raios-X (DRX), onde a fase tetragonal Nd2Fe14B foi o constituinte dominante, e por meio de análise em microscópio eletrônico de varredura (MEV). Semi-quantitativamente os ímãs, seus revestimentos e o precipitado final foram avaliados por meio de espectrometria de fluorescência de raios-X (FRX). A composição média de 21,5% de neodímio e 65,1% de ferro em massa, além de outros elementos químicos, obtida por meio de espectrometria de emissão óptica com plasma (ICP-OES), mostrou teor de neodímio superior aos minérios explorados, tornando este resíduo interessante em relação à recuperação deste metal. O procedimento hidrometalúrgico adotado envolveu a lixiviação das amostras em ácido sulfúrico 2M e a variação de três parâmetros: razão sólido/líquido, temperatura e tempo, em dois níveis. Posteriormente foi realizada a precipitação seletiva por meio de NaOH, formando o duplo sal NaNd(SO4)2.nH2O. Os resultados mostraram que a lixiviação de neodímio atingiu eficiência média acima de 94%, porém todos os experimentos apresentaram material não lixiviado, majoritariamente composto por níquel e cobre. / The chemical element neodymium (Nd) currently is in evidence as it is a fundamental part in applications manufactured by clean and advanced technology companies. However, it is classified as critical by the United States Department of Energy due to Chinese monopoly in its exploration and commercialization. Due to increasing demand in contrast with scarce supply, the recovery of neodymium is growing. One of the applications of this metal is the neodymium-iron-boron permanent magnet, used in hard disk drives (HDDs) of computers. Therefore, this study aimed magnets characterizing from HDDs of computers and then establishing a recovery route for neodymium by mechanical and hydrometallurgical processes. By weighing the HDDs and their magnets it was obtained the average weight of 3.04 g and 7.91 g for laptop and desktop, respectively, representing less than 3% of the HDDs total mass. Demagnetizing was achieved by heating up to Curie temperature of around 312 °C, and the comminution in knives mill resulted in three granulometries classified into fractions A, B and C. Qualitatively, magnets were analyzed by X-ray diffraction (XRD), where the tetragonal phase Nd2Fe14B was the dominant constituent of the sample, and through analysis in a scanning electron microscope (SEM). Semi-quantitatively, the magnets, its covers, and the final precipitate were evaluated by X-ray fluorescence spectrometry (XRF). The average composition of neodymium and iron showed 21.5% and 65.1% by weight, respectively, among other chemical elements. This results were obtained by optical emission spectrometry (ICP-OES), representing a higher neodymium content compared to the ores content, which makes this waste interesting from a neodymium recovery point of view. The procedure adopted involved hydrometallurgical leaching of the samples in 2M sulfuric acid varying three parameters: liquid/solid ratio, temperature and time, each one on two levels. Subsequently, the selective precipitation by NaOH was performed to form the double salt NaNd(SO4)2.nH2O. The results showed that neodymium leaching efficiency reached above 94%, but the experiments showed as well some material left, mainly composed of nickel and copper.

Este trabalho apresenta um estudo sobre o gargalo no acesso a disco sob arquiteturas com sistemas multiprocessadores e processadores multicore e apresenta uma proposta de solução para o problema e sua implementação. A solução desenvolvida é chamada de FPA. A estratégia desenvolvida no trabalho aproveita os recursos de memória compartilhada da arquitetura e utiliza buffers de dados compartilhados entre as diversas threads das aplicações, bem como tenta otimizar o acesso ao disco rígido. Para que seja diminuída a concorrência no acesso a disco, é permitido que apenas uma thread realize leituras ou gravações no disco por vez, e nas situações de escrita é feita a sobreposição entre o processamento e o envio dos dados ao disco. As operações de escrita são realizadas por uma thread adicional, que é inicializada e gerenciada pela FPA. Foram executados diversos testes com diferentes números de threads, tamanhos de arquivo e quantidade de processamento de cálculos na aplicação. Os resultados apresentaram ganho de desempenho com a FPA em diversas situações, onde um dos fatores mais significativos para o ganho de desempenho é a quantidade de processamento existente na aplicação. De modo geral, as operações de leitura apresentaram ganho de desempenho em relação ao uso de streams nas situações onde existe pouco processamento de cálculos na aplicação, e as operações de escrita mostraram vantagem nas situações onde existe uma quantidade de processamento de cálculos maior na aplicação. Nas operações de escrita é obtido vantagem com a sobreposição entre processamento e escrita de dados. Porém, quando a quantidade de processamento de cálculos é muito alta, o tempo de espera pela gravação em disco torna-se irrisório perante o tempo gasto no processamento dos cálculos, ocasionando perda de desempenho da FPA em relação aos streams. / This work presents a study on disk access bottlenecks for multiprocessor architecture systems and also proposes a specification to solve this problem and its implementation. The developed strategy is called FPA. The developed strategy makes use of shared memory architecture and uses shared data buffers among several application threads and also tries to optimize disk access time. In order to decrease disk concurrency, is allowed to just one application thread to read/write simultaneously and processing thread operations run parallelly to disk operations. Write operations are performed by an internal thread, which is started and managed by FPA. Several tests were performed concerning different amounts of application threads instances, file sizes and computing instructions. Results show performance gain with FPA in several situations. The main factor that affects performance is the amount of computing instructions on each application. In general, read operations showed a better performance compared to use of streams in situations where there is few computing instructions in the application, and write operations showed advantages on situations where there are more computing instructions in the application. Write operations are advantageous using data writing and computing instructions overlap. However, in cases of a large amount of computing instructions compared to disk operations, the time to perform disk operations becomes too few compared to computing time, occurring FPA performance loss compared to streams.

Stimuli responsive coatings offer a versatile method by which to manipulate interfacial interactions of proteins in a desired way. However, there exists little guidance as to how the structure of a responsive polymer coating influences adsorption of proteins. In this dissertation, the adsorption behavior of immuglobulin G (IgG) on poly (N-isopropylacryamide) (PNIPAAm) hydrogel coatings was investigated as a function of film thickness. PNIPAAm exhibits a hydrophilic to hydrophobic transition above a critical temperature of ~32°C in aqueous solutions. In this research, through the use of quartz crystal microbalance with dissipation (QCM-D) it was observed that the adsorption was thickness dependent and became non-reversible as the temperature was decreased. Interestingly, QCM-D results also suggested a similar amount of protein adsorption on both hydrated and dehydrated PNIPAAm surfaces. A rigid film analysis using Sauerbrey equation revealed a multi-layer formation on the collapsed PNIPAAm coatings. Although it is allegedly reported that PNIPAAm favors adsorption above the critical temperature due to hydrophobic interactions, there have been several studies that reported adsorption of proteins below the critical temperature. To better understand the QCM-D results, hydrodynamic shear force assays in a spinning disk configuration were performed in order to quickly measure and quantify adhesion of polystyrene (PS) probe spheres (10μm) to the PNIPAAm coatings in both the solvated (hydrophilic) and collapsed (hydrophobic) state. The influence of polymer coating thickness, polymer chain cross-link density, microsphere concentration and adsorption time on the adhesion characteristics of the coatings was investigated in relation with volume phase transition of the polymer coatings. A series of experiments on quantification of the temperature dependent adhesion of proteins adsorbed on surface attached PNIPAAm coatings of thicknesses was performed as the surface chemistry was switched from hydrophilic to hydrophobic. First, adhesion of polystyrene (PS) microspheres on PNIPAAm coatings was quantified in order to have a guideline for temperature dependent adhesion performance of these coatings. PS particles were subjected to a range of detachment shear stresses through hydrodynamic flow in a spinning disk configuration. These experiments provide an indirect method to determine the force of adhesion since it is proportional to the hydrodynamic force. Model protein, IgG, was then linked to PS microspheres and the mean adhesion strength of the IgG coated PS microspheres were determined through the detachment shear stresses. The influence of PS deposition time, PS bead concentration, PNIPAAm coating thickness and PNIPAAm cross-link density on the adhesion strength were addressed. The results indicated that in the collapsed state, the adhesion of bare hydrophobic PS microspheres depends strongly on coating thickness. For hydrophilic charged PS microspheres the adhesion was always higher on the hydrated PNIPAAm surfaces and appeared not to be strongly affected by the increase in PNIPAAm coating thickness. The adhesion of IgG was higher on the collapsed PNIPAAm surfaces and the adhesion trend did not significantly change as the PNIPAAm film thickness was increased. For PNIPAAm coatings with the cross-link density reduced by factor of 10, the adhesion was again higher on the collapsed PNIPAAm surface and scaled linearly with thickness. Moreover, the influence of thickness became prominent at the higher thickness values (165 nm-185 nm). In addition, the adhesion of carboxylated microspheres on PNIPAAm did not reach equilibrium and increased linearly with microsphere deposition time. A study on the sensing characteristics of PNIPAAm coatings in response to heavy metal ions was also conducted in this dissertation. The temperature-dependent swelling behavior of poly(N-isopropylacrylamide) and tripeptide Gly-Gly-His/poly(NIPAAm) conjugate hydrogel coatings were investigated using a quartz crystal microbalance with dissipation (QCM-D) while in contact with NaCl, ZnCl2, NiCl2, and CuCl2 solutions. To fabricate the tripeptide conjugated gels, precursor gels of poly(NIPAAm-co-3-aminopropylmethacrylamide[3.5 mole%]) were synthesized via free radical polymerization. The metal binding tripeptide, Gly-Gly-His, was subsequently synthesized in the gel via a Merrifield solid phase peptide synthesis (SPPS) technique, in which the amino group of the copolymer gel provided a functional site to support peptide synthesis. It was found that the logarithm of the transition temperature of the tripeptide Gly-Gly-His/poly(NIPAAm) conjugate hydrogel was proportional to the ionic strength, showing two distinct regions at low and high ionic strengths for the divalent ions. In the low ionic strength regime, the salting out constants were 0.08 M-1, 0.07 M-1, and 0.06 M-1 for Cu2+, Ni2+, and Zn2+, respectively, which follows the known trend for binding of the ions to Gly-Gly-His. In the high ionic strength region, when the metal-ion binding sites in the tripeptide conjugate hydrogel were saturated, the salting out constants were similar to the salting out constants associated with pure poly(NIPAAm).

Bibliography: leaves 107-109 / 109 leaves, [26] leaves of plates : / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (M.D.)--University of Adelaide, 1989

Degeneration of the lumbar intervertebral disc, a condition widely implicated in the cause of low back pain among adult humans, is typically characterised by progressive biochemical and structural changes to the extracellular matrix. Comprehensive descriptions of the structural and functional inter-relationships within the extracellular matrix are therefore critical to understanding the degenerative process and developing effective treatments. In the anulus fibrosus, this matrix has a complex, hierarchical architecture comprised of collagens, proteoglycans, and elastic fibres. Elastic fibres are critical constituents of dynamic biological structures that functionally require elasticity and resilience. Studies to date of elastic fibre network structure in the anulus fibrosus have been qualitative and limited in scope. Additionally, there is poor understanding of the structural and functional associations between elastic fibres and other matrix constituents such as collagen, and, critically, there have been no studies directly examining the nature and magnitude of the contribution made by elastic fibres to anulus fibrosus mechanical behaviour. In this thesis, multiple experimental studies are described that specifically examine each of these areas. Novel imaging techniques were developed and combined with histochemistry and light microscopy to facilitate the visualisation of elastic fibres at a level of detail not previously achieved. Examination of elastic fibre network structure revealed architectural differences between the intralamellar and interlamellar regions, suggesting that elastic fibres perform functional roles at distinct levels of the anulus fibrosus structural hierarchy. The density of elastic fibres within lamellae was found to be significantly higher in the lamellae of the posterolateral region of the anulus than the anterolateral, and significantly higher in the outer regions than the inner, suggesting it may be commensurate with the magnitude of the tensile strains experienced by each region of the disc in bending and torsion. The nature of the structure-function associations between elastic fibres and collagen was then examined with respect to the reported structural mechanisms of collagen matrix tensile deformation. Histological assessment of collagen crimp morphology in specimens from which elastic fibres had been enzymatically removed revealed no observable differences when compared with controls, suggesting that any contribution made by elastic fibres to maintaining crimp is minimal. Elastic fibres in anulus fibrosus specimens subjected to radial tensile deformations exhibited complex patterns of re-arrangement, suggesting that they maintain cross-collagen fibre connectivity. Elastic fibres were also observed to maintain physical connections between consecutive lamellae undergoing relative separation. Finally, the nature and magnitude of the contribution made by elastic fibres to anulus fibrosus mechanical properties at the tissue level was investigated using a combination of biochemically verified enzymatic treatments and biomechanical tests. Targeted degradation of elastic fibres resulted in a significant reduction in both the initial modulus and the ultimate modulus, and a significant increase in the extensibility, of radially oriented anulus fibrosus specimens. Separate treatments and mechanical tests were used to account for any changes attributable to non-specific degradation of glycosaminoglycans. These results suggest that elastic fibres enhance the mechanical integrity of the anulus fibrosus extracellular matrix in the direction perpendicular to the plane containing the collagen fibres. In summary, the results of the studies presented in this thesis provide important new insights into the structure and function of the anulus fibrosus elastic fibre network, and highlight its potential importance as a contributing or ameliorating factor in the progression of the structural and mechanical changes associated with intervertebral disc degeneration. Additionally, these results establish an improved framework for the development of more accurate analytical and finite element models to describe disc behaviour. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1317006 / Thesis (Ph.D.) -- University of Adelaide, School of Medical Sciences, 2008

One of the most important problems that plague turbomachinery rotors is the existence of rogue blades -- lone blades that exhibit unexpected fatigue failure. It has been recognized that rotor mistuning might be the cause of rogue blades through a phenomenoncalled normal mode localization, whereby vibration energy is confined to a few blades of the assembly. The goals of this dissertation are (1) to achieve a thorough understanding of the fundamental mechanisms governing mistuning effects, (2) the development of mathematical models of turbomachinery rotors suitable for mistuning analysis, and (3) the development of techniques for designers interested in the mistuning sensitivity of a particular rotor design.