Over the past few decades, we have been enjoying tremendous benefits thanks to the revolutionary advancement of computing systems, driven mainly by the remarkable semiconductor technology scaling and the increasingly complicated processor architecture. However, the exponentially increased transistor density has directly led to exponentially increased power consumption and dramatically elevated system temperature, which not only adversely impacts the system's cost, performance and reliability, but also increases the leakage and thus the overall power consumption. Today, the power and thermal issues have posed enormous challenges and threaten to slow down the continuous evolvement of computer technology. Effective power/thermal-aware design techniques are urgently demanded, at all design abstraction levels, from the circuit-level, the logic-level, to the architectural-level and the system-level. In this dissertation, we present our research efforts to employ real-time scheduling techniques to solve the resource-constrained power/thermal-aware, design-optimization problems. In our research, we developed a set of simple yet accurate system-level models to capture the processor's thermal dynamic as well as the interdependency of leakage power consumption, temperature, and supply voltage. Based on these models, we investigated the fundamental principles in power/thermal-aware scheduling, and developed real-time scheduling techniques targeting at a variety of design objectives, including peak temperature minimization, overall energy reduction, and performance maximization. The novelty of this work is that we integrate the cutting-edge research on power and thermal at the circuit and architectural-level into a set of accurate yet simplified system-level models, and are able to conduct system-level analysis and design based on these models. The theoretical study in this work serves as a solid foundation for the guidance of the power/thermal-aware scheduling algorithms development in practical computing systems.

Understanding the mechanical properties and deformation characteristics of nanoscale metallic nanopillars and wires is a significant concern for designing reliable small devices that must resist loads in service. This thesis aims to extend understanding of the size dependent behaviour of nanopillars and wires in compression and tension by investigating their mechanical properties and deformation characteristics. Single crystal bcc pillars were fabricated by focussed ion beam (FIB) machining from Fe, Nb, V, Ta, Mo, W and Cr, as well as the ferrite (bcc) and austenite (fcc) components of a duplex stainless steel (DSS). These were tested in compression over a range of test temperatures from 193 K to 393 K using various types of nanomechanical devices. The effect of sample size (pillar diameter) on the strength was investigated and found to increase with decreasing pillar size. In bcc metals, the yield or flow stress, 􀀂􀀖, is inversely proportional with some power of the pillar diameter, d. In bcc metals tested, the power-law exponent, n, were found in the range of between -0.23 to -0.63, showing a less pronounced size effect than found for fcc pillars. The power-law exponent for bcc pillar deformation is also temperature dependent and was found to scale with the ratio of test temperature, Ttest to the critical temperature for screw dislocation mobility, Tc, of the bcc metal (T*= Ttest / Tc). It is notable that the size effect exponent weakens (approaches 0) as T* decreases. However, when the experiments are carried out at temperatures close to or just above Tc, the power-law exponents approaches the value reported in the literature for a range of fcc metals (-1 < n < -0.6). The variation in the power-law exponent observed for bcc metals can be explained by the change in mobility of thermally activated screw dislocations. Their mobility can be modelled by a threshold or lattice friction stress. If this friction stress is introduced into the empirical equation that relates the strength of fcc metal pillars to their diameter, a strong correlation between size effect exponent, the normalised test temperature (T*) and friction stress is obtained. It was found that the friction stress values (Fe, Nb and V) increase as Ttest decreases from 296 to 193 K. When the pillar diameter decreases, the friction stress would be more easily overcome due to the increase in surface-to-volume ratio. The contribution of lattice friction stress on the strength is higher at larger pillars than those for nanopillars. Thus, the divergence between best fit lines has become more apparent at micron-sized pillars, resulting in weaker size effects. Furthermore, the transition in deformation morphology from localized to wavy deformation was only found in Fe pillars, as the Ttest decreased from 296 to 193 K, further revealing that temperature has also strong influence on deformation behaviours of bcc pillars.

Les matériaux composites à matrice organique sont largement utilisés dans l'industrie des transports et notamment dans le domaine aéronautique. Pour permettre un dimensionnement optimal des structures, il est nécessaire d'étudier le comportement des matériaux CMO sur une large gamme de vitesses et de températures.L'objectif de cette thèse est de proposer un modèle de comportement et de rupture permettant de prédire la réponse des CMO sur une large gamme de vitesses de sollicitation et de températures. Les recherches se sont intéressées dans un premier temps à la caractérisation de la transition entre les régimes de comportement linéaire et non linéaire du matériau unidirectionnel T700GC/M21 (renforts de fibres de carbone, résine époxy), ainsi qu'à la dépendance de cette transition à la vitesse de sollicitation et à la température. Les travaux se sont ensuite focalisés sur l'étude expérimentale du régime de comportement non linéaire endommageable du T700GC/M21. Enfin, au terme de ces deux étapes, une version améliorée du modèle disponible à l'ONERA pour les composites stratifiés (OPFM) a été proposée, version intégrant un critère de transition linéaire/non linéaire de comportement, et une prise en compte de l'influence de la vitesse de sollicitation et de la température sur la réponse du matériau / Nowadays, organic matrix composite materials are widely used in the transportation industry, and particularly in the aeronautical industry. To provide an optimal dimensioning of the structures, it is necessary to study the mechanical behavior of OMC on a large range of strain rates and temperatures. The aim of this PhD thesis is to propose a behavior and a rupture model to predict the mechanical response of OMC for a large range of strain rates and temperatures. The research was initially focused on the characterization of the transition between the linear and nonlinear behavior of the material T700GC/M21, a carbon / epoxy unidirectional laminate as well as the strain rate and temperature dependencies of this transition. The work was then focused on the experimental study of the nonlinear damaged behavior of the T700GC/M21. Finally, completing these first two steps, an updated version of the behavior model available at ONERA (OPFM) was proposed which includes the transition between linear and nonlinear behavior and the influence of strain rate and temperature on the mechanical response of the material.