Le but du travail réalisé dans cette thèse est de fournir aux fournisseurs de services une solution qui est capable de non seulement déployer les services dans le cloud de façon économique, automatique, mais aussi à grande échelle. La première contribution traite le problème de la construction d’un nouveau service demandé par le client à partir de services disponibles dans le cloud de manière à satisfaire les exigences en termes de qualité de service ainsi qu’en termes de coût. Nous présentons ce problème par un modèle analytique et proposons un algorithme heuristique dont la performance est améliorée de 20-30% par rapport aux autres approches. La seconde contribution est une solution pour déployer les services dans le cloud en considérant les demandes des utilisateurs finaux. Pour assurer qualité de services dans une grande échelle, le service demandé est dupliqué et distribué dans le réseau; chacun de ses réplicas servira les utilisateurs à proximité. Le plan d’approvisionnement selon lequel le service est dupliqué dépend de sa demande, ce qui ne cesse pas de changer en quantité ainsi qu’en distribution, ce qui rend le problème plus compliqué. Nous proposons une solution qui est capable de s’adapter aux changements dans le réseau, y compris ceux des demandes de ses utilisateurs. Enfin, nous proposons un système basé sur OpenStack qui permet de déployer les services complexes dans un cloud qui couvre différente locations (cloud multi-site). A partir d’une demande du client, le système automatiquement calcule le plan d’approvisionnement optimal et le déploie en respectant les contraintes du client. / The purpose of the work in this thesis is to provide the Service Provider a solution which is capable of deploying complex services in a cloud automatically and cost-effectively. The first contribution allows the Service Provider to construct complex services requested by the clients from basic services at his disposal. The construction must be efficient in terms of execution time and operation cost while respecting the client’s constraints. We present an analytic model for this problem and propose a heuristic solution which performance is 20-30% better than other approaches. The second contribution solves the problem of deploying the services while considering end-users’ demands. To ensure the quality of services provided to end-users, not only one instance but a set of service replicas is deployed in the network. How the service is duplicated and distributed depends on the demands of its end-users that change constantly in quantity as well as distribution, which complicates the problem. Thus the provisioning mechanism has to be capable of adapt to the change in the network, including the change in end-users’ demands. Our third contribution is a system using OpenStack which allows Service Provider to deploy complex services in a cloud that spans over different locations (multi-site cloud). Given a client’s request, the system automatically calculates the optimal provisioning plan and deploys it while respecting the client’s constraints.

Caches are known to consume up to half of all system power in embedded processors. Co-optimizing performance and power of the cache subsystems is therefore an important step in the design of embedded systems, especially those employing application specific instruction processors. In this project, we propose an analytical cache model that succinctly captures the miss performance of an application over the entire cache parameter space. Unlike exhaustive trace driven simulation, our model requires that the program be simulated once so that a few key characteristics can be obtained. Using these application-dependent characteristics, the model can span the entire cache parameter space consisting of cache sizes, associativity and cache block sizes. In our unified model, we are able to cater for direct-mapped, set and fully associative instruction, data and unified caches. Validation against full trace-driven simulations shows that our model has a high degree of fidelity. Finally, we show how the model can be coupled with a power model for caches such that one can very quickly decide on pareto-optimal performance-power design points for rapid design space exploration. / Singapore-MIT Alliance (SMA)

The energy requirements of forage harvesters can be quite high and can sometimes determine the size of tractor needed on a farm. Therefore, improving the energy efficiency of the forage harvester could allow a farm to reduce costs by using a smaller tractor that is less expensive and more efficient. The objective of this research was to increase the throwing distance of a forage harvester by modeling the flow of forage in the spout and the air flow in the blower and spout. These models can then be used to compare the efficiencies of prototype designs. The air flow in the blower and spout was modeled using the commercial computational fluid dynamics software FLUENT. The simulation results of air velocities and flow patterns were compared to experimental values and it was found that both were of the same order of magnitude with the model predicting slightly higher air velocities than those measured. The flow of forage in the spout was modeled analytically by taking into account the friction between the forage and the spout surface and the aerodynamic resistance after the forage leaves the spout. From this model, two improved prototype spouts that should theoretically result in longer throwing distances were designed. However, field testing of the two prototypes did not reveal any significant improvements over the current design. It was also found that the model under-predicted the throwing distance of one prototype by 2 % and over estimated the other by 12 %.

The energy requirements of forage harvesters can be quite high and can sometimes determine the size of tractor needed on a farm. Therefore, improving the energy efficiency of the forage harvester could allow a farm to reduce costs by using a smaller tractor that is less expensive and more efficient. The objective of this research was to increase the throwing distance of a forage harvester by modeling the flow of forage in the spout and the air flow in the blower and spout. These models can then be used to compare the efficiencies of prototype designs. The air flow in the blower and spout was modeled using the commercial computational fluid dynamics software FLUENT. The simulation results of air velocities and flow patterns were compared to experimental values and it was found that both were of the same order of magnitude with the model predicting slightly higher air velocities than those measured. The flow of forage in the spout was modeled analytically by taking into account the friction between the forage and the spout surface and the aerodynamic resistance after the forage leaves the spout. From this model, two improved prototype spouts that should theoretically result in longer throwing distances were designed. However, field testing of the two prototypes did not reveal any significant improvements over the current design. It was also found that the model under-predicted the throwing distance of one prototype by 2 % and over estimated the other by 12 %.

Recently, experimental observations have shown that radar echoes from the irregularity<br />source region associated with mesospheric dusty space plasmas may be modulated by radio wave heating with ground-based ionospheric heating facilities. These experiments show great promise as a diagnostic for the associated dusty plasma in the Near-Earth Space Environment which is believed to have links to global change. This provides an alternative to more complicated and costly space-based observational approaches to investigating these layers. This dissertation seeks to develop new analytical and computational models to investigate fundamental physics of the associated dusty plasmas as well as utilize experimental observations during High Frequency HF ground-based heating experiments to develop practical techniques for diagnosing these dusty plasma layers.<br />The dependency of the backscattered signal strength (i.e. Polar Mesospheric Summer Echoes PMSEs) after the turn-on and turn-off of the radio wave heating on the radar frequency is an unique phenomenon that can shed light on the unresolved issues associated with the basic physics of the natural charged mesospheric dust layer. The physical process after turn-on and turn-off of radio wave heating is explained by competing ambipolar diffusion and dust charging processes. The threshold radar frequency and dust parameters for the enhancement or suppression of radar echoes after radio wave heating turn-on are investigated for measured mesospheric plasma parameters. The effect of parameters such as the electron temperature enhancement during radiowave heating, dust density, dust charge polarity, ion-neutral collision frequency, electron density and dust radius<br />on the temporal evolution of electron irregularities associated with PMSE is investigated.<br />The possibility of observing the turn-on overshoot (enhancement of radar echoes after the<br />radiowave turn-on) in the high frequency HF radar band is discussed based on typical mesospheric<br />parameters. It has been shown that predicted enhancement of electron irregularity<br />amplitude after heater turn-on at HF band is the direct manifestation of the dust charging<br />process in the space. Therefore further active experiments of PMSEs should be pursued<br />at HF band to illuminate the fundamental charging physics in the space environment to<br />provide more insight on this unique medium. Preliminary observation results of HF PMSE<br />heating experiment with the new 7.9 MHz radar at the European Incoherent Scatter EISCAT<br />facility appear promising for the existence of PMSE turn-on overshoot. Therefore, future<br />experimental campaigns are planned to validate these predictions.<br />Computational results are used to make predictions for PMSE active modification experiments at 7.9, 56, 139, 224 and 930MHz corresponding to existing ionospheric heating facilities. Data from a 2009 very high frequency VHF (224 MHz) experiment at EISCAT<br />is compared with the computational model to obtain dust parameters in the PMSE. The<br />estimated dust parameters as a result of these comparison show very reasonable agreement to dust radius and density at PMSE altitudes measured during a recent rocket experiment providing validation to the computational model.<br /><br />The first comprehensive analytical model for the temporal evolution of PMSE after heater<br />turn-on is developed and compared to a more accurate computational model as a reference.<br />It is shown that active PMSE heating experiments involving multiple observing frequencies<br />at 7.9 (HF), 56, and 224 MHz (VHF) may contribute further diagnostic capabilities since<br />the temporal evolution of radar echoes is substantially different for these frequency ranges.<br />It is shown that conducting PMSE active experiments at HF and VHF band simultaneously<br />may allow estimation of the dust density altitude profile, dust charge state variation during<br />the heating cycle, and ratio of electron temperature enhancement in the irregularity source<br />region. These theoretical and computational models are extended to study basic physics of the evolution of relevant dusty plasma instabilities thought to play an important role in irregularity production in mesospheric dust layers. A key focus is the boundary layer of these charged dust clouds. Several aspects of the cloud\'s structure (thickness of boundary layer, average particle size and density, collisional processes, and cloud expansion speed) and the ambient plasma are varied to determine the effect of these quantities on the resulting irregularities.<br />It was shown that for high collision frequencies, the waves may be very weakly excited (or<br />even quenched) and confined to the boundary layer. The excited dust acoustic waves inside<br />the dust cloud with frequency range of 7-15Hz and in the presence of electron bite-outs is<br />consistent with measured low frequency waves near 10 Hz by sounding rocket experiments<br />over the past decade. The observed radar echoes associated with the artificially created dust<br />clouds at higher altitudes in the ionosphere including space shuttle exhaust and upcoming<br />active space experiments in which localized dust layers will be created by sounding rockets<br />could be related to the excited acoustic waves predicted.<br />Finally, variation of spatial structures of plasma and dust (ice) irregularities in the PMSE<br />source region in the presence of positively charged dust particles is investigated. The correlation and anti-correlation of fluctuations in the electron and ion densities in the background plasma are studied considering the presence of positive dust particle formation. Recent rocket payloads have studied the properties of aerosol particles within the ambient plasma environment in the polar mesopause region and measured the signature of the positively charged particles with number densities of (2000 cm"3) for particles of 0.5-1 nm in radius.<br />The measurement of significant numbers of positively charged aerosol particles is unexpected from the standard theory of aerosol charging in plasma. Nucleation on the cluster ions is one of the most probable hypotheses for the positive charge on the smallest particles. The utility being that it may provide a test for determining the presence of positive dust particles.<br />The results of the model described show good agreement with observed rocket data. As an<br />application, the model is also applied to investigate the electron irregularity behavior during<br />radiowave heating assuming the presence of positive dust particles. It is shown that the<br />positive dust produces important changes in the behavior during Polar Mesospheric Summer Echo PMSE heating experiments that can be described by the fluctuation correlation and anti-correlation properties.<br />The second part of this dissertation is dedicated to Stimulated Electromagnetic Emissions SEEs produced by interaction of high power electromagnetic waves in the ionosphere. Nearearth ionospheric plasma presets a neutral laboratory for investigation of nonlinear wave phenomena in plasma which can not be studied in the laboratory environment due to the effect of physical boundary conditions. This process has been of great interest due to the<br />important diagnostic possibilities involving ability to determine mass of constitutive ions in<br />the interaction region through measurements of various gyro-frequencies. Objectives include<br />the consideration of the variation of the spectral behavior under pump power, proximity to<br />the gyro-harmonic frequency, and beam angle. Also, the relationship between such spectral<br />features and electron acceleration and creation of plasma irregularities was an important<br />focus.<br />Secondary electromagnetic waves excited by high power electromagnetic waves transmitted<br />into the ionosphere, commonly know as Stimulated Electromagnetic Emissions SEEs,<br />produced through Magnetized Stimulated Brillouin Scatter MSBS are investigated. Data<br />from two recent research campaigns at the High Frequency Active Auroral Research Program<br />facility HAARP is presented in this work. These experiments have provided additional<br />quantitative interpretation of the SEE spectrum produced by MSBS to yield diagnostic measurements of the electron temperature in the heated ionosphere. SEE spectral emission lines corresponding to ion acoustic IA and electrostatic ion cyclotron EIC modes were observed with a shift in frequency up to a few tens of Hz from radio waves transmitted near the third harmonic of the electron gyro-frequency 3fce. The threshold of each emission line has been measured by changing the pump wave amplitude. The experimental results aimed to show the threshold for transmitter power to excite IA waves propagating along the magnetic field lines as well as for EIC waves excited at oblique angles relative to the background magnetic field. A full wave solution has been used to estimate the amplitude of the electric field at the interaction altitude. The estimated growth rate using the theoretical model is compared with the threshold of MSBS lines in the experiment and possible diagnostic information for the background ionospheric plasmas is discussed. Simultaneous formation of artificial field aligned irregularities FAIs and suppression of the MSBS process is investigated. Recently, there has been significant interest in ion gyro-harmonic structuring the Stimulated Electromagnetic Emission SEE spectrum due to the potential for new diagnostic information available about the heated volume and ancillary processes such as creation of artificial ionization layers. These relatively recently discovered emission lines have almost exclusively been studied for second electron gyro-harmonic heating. The first extensive systematic investigations of the possibility of these spectral features for third electron gyro-harmonic heating are provided here. Discrete spectral features shifted from the transmit frequency ordered by harmonics of the ion gyro-frequency were observed for third electron gyro-harmonic heating for the first time at a recent campaign at a High Frequency Active Auroral Research Program Facility HAARP. These features were also closely correlated with a broader band feature at a larger frequency shift from the transmit frequency known as the Downshifted Peak DP. The power threshold of these spectral features was measured, as well as their behavior with heater<br />beam angle, and proximity of the transmit frequency to the third electron gyro-harmonic frequency. Comparisons were also made with similar spectral features observed during 2nd<br />electron gyro-harmonic heating during the same campaign. A theoretical model is provided<br />that interprets these spectral features as resulting from parametric decay instabilities in<br />which the pump field ultimately decays into high frequency upper hybrid/electron Bernstein<br />and low frequency neutralized ion Bernstein IB and/or obliquely propagating ion acoustic<br />waves at the upper hybrid interaction altitude. Coordinated optical and SEE observations<br />were carried out in order to provide a better understanding of electron acceleration and precipitation<br />processes. Optical emissions were observed associated with SEE gyro-harmonic<br />features for pump heating near the second electron gyro-harmonic during the campaign. The<br />observations affirm strong correlation between the gyro-structures and the airglow. / Ph. D.

The cold-formed steel framed shear walls with steel sheets and wood-based sheathing are both code approved lateral force resisting system in light-framed construction. In the United States, the current design approach for cold-formed steel shear walls is capacity-based and developed from full-scale tests. The available design provisions provide nominal shear strength for only limited wall configurations. This research focused on the development of analytical models of cold-formed steel framed shear walls with steel sheet and wood-based sheathing to predict the nominal shear strength of the walls at their ultimate capacity level. Effective strip model was developed to predict the nominal shear strength of cold-formed steel framed steel sheet shear walls. The proposed design approach is based on a tension field action of the sheathing, shear capacity of sheathing-to-framing fastener connections, fastener spacing, wall aspect ratio, and material properties. A total of 142 full scale test data was used to verify the proposed design method and the supporting design equations. The proposed design approach shows consistent agreement with the test results and the AISI published nominal strength values. Simplified nominal strength model was developed to predict the nominal shear strength of cold-formed steel framed wood-based panel shear walls. The nominal shear strength is determined based on the shear capacity of individual sheathing-to-framing connections, wall height, and locations of sheathing-to-framing fasteners. The proposed design approach shows a good agreement with 179 full scale shear wall test data. This analytical method requires some efforts in testing of sheathing-to-framing connections to determine their ultimate shear capacity. However, if appropriate sheathing-to-framing connection capacities are provided, the proposed design method provides designers with an analytical tool to determine the nominal strength of the shear walls without conducting full-scale tests.

A temperatura do solo influencia a maioria dos processos físicos, químicos e biológicos que ocorrem no solo. O modelo analítico exponencial-senoidal em uma dimensão descreve razoavelmente bem a temperatura no solo, T (oC), como função do tempo, t (s), e da profundidade, z (m), 0 &#8804; z < &#8734;. A temperatura da superfície pode ser representada pela soma de duas senóides, uma relacionada com variações de temperatura anuais, outra com variações diárias, cada uma tendo uma amplitude constante. Uma correção para a variação temporal de amplitude diária é aqui introduzida. A equação do calor é resolvida analiticamente, com pouco aumento em complexidade em relação à solução tradicional. Predições de temperatura obtidas pela nova solução analítica foram comparadas com predições obtidas da solução usual, que trata a amplitude diária como constante. Para comparar as predições, foram necessários valores experimentais de certos parâmetros que aparecem nestas equações; foi suficiente usar valores típicos, obtidos na literatura. Predições são comparadas utilizando quatro conceitos: (i) profundidade de amortecimento, D; (ii) profundidade de penetração, zM; (iii) erro quadrático médio (EQM); e (iv) erro quadrático médio na forma de uma integral (EQMI). O conceito de zM foi aqui introduzido, acompanhado por uma equação simples que permite calcular qual é a profundidade zm tal que, se temperatura T(zM,t) for aproximada como Ta (valor médio da temperatura ao longo do ano, em zM), o erro em tal aproximação será igual ou menor um certo valor previamente definido, por exemplo, de 0,1 oC. O conceito de EQMI, também introduzido nesta tese, substitui o somatório que aparece no EQM por uma integral definida, e serve para comparar dois modelos analíticos, o que era o caso desta tese. Valores de D e zM mostram que a correção é desprezível para z > 0,6 m. Valores de EQM mostram que a correção é considerável para z = 0,1 m. Nesta profundidade, o valor máximo foi EQM = 0.30 oC para dias inteiros, e EQM = 0.29 oC para meses inteiros. Valores de EQMI foram praticamente iguais aos valores de EQM. Para qualquer profundidade a correção introduzida, ainda que considerável, é pequena. Entretanto, a única informação adicional requerida para aplicar a equação nova é a informação de variação temporal da amplitude diária. Desta forma, pode-se sugerir que a nova equação seja preferida, quando esta informação esteja facilmente disponível a partir de dados experimentais. / Soil temperature influences many physical, chemical and biological processes that occur in soil. The exponential-sinusoidal one-dimensional analytical model reasonably describes soil temperature, T (oC), as a function of time, t (s), and depth, z (m), 0 &#8804; z < &#8734;. Surface temperature may be represented by the sum of two sinusoids, one related to annual and the other to daily temperature variations, each one having constant amplitude. A correction for the temporal variation of daily amplitude is introduced here. The heat equation is solved analytically, with minimal increase in complexity compared to the traditional solution. Temperature predictions obtained from the novel analytical solution are compared with predictions from the usual solution that treats the daily amplitude as a constant. Comparisons demanded experimental parameters, which were obtained from scientific literature. Predictions are compared using four concepts: (i) damping depth, D; (ii) penetration depth, zM; (iii) root mean squared error (RMSE); and (iv) root mean squared error defined by a definite integral (RMSEI). The concept of zM was introduced here, through a simple equation, which allows calculation of the depth zm, at which T(zM,t) can be approximated to Ta (average annual value of soil temperature at zM). The concept of RMSEI was also introduced here, and replaces a sum by a definite integral. The RMSEI can be used to compare analytical models, as it was the case here. Values of D and zM show that the correction is negligible for z > 0,6 m. Values of RMSE show that the correction is considerable for z = 0,1 m. For individual days, at a depth z = 0,1 m, the maximum value was RMSE = 0.30 oC; for whole months, the maximum value was RMSE = 0.29 oC. RMSEI values were practically the same as RMSE values. The correction introduced here was small at all depths. However, the only additional information required to apply the novel equation is information on temporal variation of daily amplitude, so this equation should be preferred when such data are readily available.

We review MIT research in manufacturing systems engineering, and we describe current and possible future research activities in this area. This includes advances in decomposition techniques, optimization, token-based control systems analysis, multiple part types, inspection location, data collection and several other topics. / Singapore-MIT Alliance (SMA)

We present methods of calculating the value of two performance parameters for multipath, multistage interconnection networks: the normalized throughput and the probability of successful message transmission. We develop a set of exact equations for the loading probability mass functions of network channels and a program for solving them exactly. We also develop a Monte Carlo method for approxmiate solution of the equations, and show that the resulting approximation method will always calculate the values of the performance parameters more quickly than direct simulation.

The mobility allowance shuttle transit (MAST) system is a hybrid transit system in which vehicles are allowed to deviate from a fixed route to serve flexible demand. A mixed integer programming (MIP) formulation for the static scheduling problem of a multi-vehicle Mobility Allowance Shuttle Transit (MAST) system is proposed in this thesis. Based on the MIP formulation, we analyze the impacts of time headways between consecutive transit vehicles on the performance of a two-vehicle MAST system. An analytical framework is then developed to model the performance of both one-vehicle and two-vehicle MAST systems, which is used to identify the critical demand level at which an increase of the fleet size from one to two vehicles would be appropriate. Finally, a sensitivity analysis is conducted to find out the impact of a key modeling parameter, w1, the weight of operations cost on the critical demand. In this paper, we develop an insertion heuristic for a multi-vehicle MAST system, which has never been addressed in the literature. The proposed heuristic is validated and evaluated by a set of simulations performed at different demand levels and with different control parameters. By comparing its performance versus the optimal solutions, the effectiveness of the heuristic is confirmed. Compared to its single-vehicle counterpart, the multiple-vehicle MAST prevails in terms of rejection rate, passenger waiting time and overall objective function, among other performance indices.