With rapid adoption of cloud solutions across industries, energy consumed by server farms continues to rise. There are numerous approaches to reduce energy consumption in data centres, and one of the approaches is to use energy-aware policies, which focus on how servers should be operated in order to achieve energy saving and meet service level agreements (SLA). In this thesis, we focus on studying a single server model with dynamic voltage scaling (DVS), presenting a framework with explicit solutions to solve for performance metrics and energy consumption. Our framework is convenient and in- tuitive, one can easily identify expected response time and expected energy consumption for a given policy. In addition, we also provide insights on how the value of the faster service rate and the choice of when to use speed scaling impact energy consumption and performance metrics. / Thesis / Master of Computer Science (MCS)

No description available.

Acid fracturing stimulation is one of the preferred methods to improve well productivity in carbonate reservoirs. Acid is injected into the fractured zone after a starter fracture is created in the near wellbore area by viscous fluid (pad). This results in propagation of a two-wing crack away from the perforations with simultaneous dissolution etching of the created surfaces. If the created etched surface is non-uniform, then after the treatment ends and the fracture face closes, a high conductivity path may remain in the formation, connected to the well. The important factors controlling the effectiveness of acid fracturing are the etched-fracture penetration and conductivity. In this research, I use the distributed volumetric sources (DVS) method to calculate gas production from a well stimulated by acid fracturing. The novel concept realized in this research is that, during the production process, the conductivity of the acid created fracture changes. I use the Nierode - Kruk correlation to describe this effect as a function of effective closure stress that in turn is determined from the flowing bottomhole pressure and minimum horizontal stress. By combining the well productivity calculation from the DVS method taking into account varying fracture conductivity with gas material balance, I obtain an improved model of gas production. The model is then used to not only forecast production from acid fractured wells but also to evaluate the known production history of such wells. Based on the concepts discussed above, I have developed a program called "Gas Acid" which is useful to optimize acid fracturing treatments and also suitable to infer created fracture parameters from known production history. The "Gas Acid" program has been validated with data from two Saudi Aramco gas wells. It was found that the production forecast obtained from the "Gas Acid" program matches the actual production history with reasonable accuracy and the remaining discrepancy could be resolved by taking into account refinement of the material balance. The refinement became necessary, because the "Gas Acid" program was developed for dry gas but the reservoir fluids in the field examples were classified as retrograde gas and wet gas. When accounting for the additional mass of gas "hidden" in the produced condensate, the match of forecast and actual data was improved considerably.

Productivity index (PI) is an important indicator of a well?s production capacity. For conventional reservoirs, well productivity is usually calculated using the pressure response of the reservoir in its pseudosteady-state period. There are numerous studies for different well completion schemes which developed correlations for pseudosteady-state productivity index for specific cases, such as horizontal wells and fractured wells. Most of the developed models for complex well completion schemes use some approximations for productivity index calculation and they have some limitations in use. Furthermore, as the petroleum industry goes toward producing lower quality reservoirs like low- and ultra low-permeability reservoirs, the period of transient flow covers a larger part of the well lifetime and these pseudosteady-state productivity calculations become less applicable in prediction of the reservoir?s production behavior. The Distributed Volumetric Sources (DVS) method seems able to fill this gap. Our method is able to predict the productivity index of a general well completion scheme for transient as well as pseudosteady-state flow periods. In this study, we focus on a typical well completion scheme ? vertical well intersected by a vertical fracture of finite conductivity. Parametric study is performed by varying the proppant pack permeability with a linear distribution, varying fracture width with an elliptical distribution and varying fracture height with an elliptical distribution. The details of hydraulic fracture are integrated into the calculation of well productivity. By combining the well productivity with gas material balance, production forecasting of the hydraulically fractured wells could be easily obtained. The result of production forecasting could be used to aid in decision making of choosing the best stimulation treatment. Field examples are presented to illustrate the application of this technology for production modeling the complicated reservoir cases involving fracture stimulation.

This dissertation provides a comprehensive overview of the theory and practice of Dynamic Voltage/Frequency Scaling (DVFS) in the High Performance Computing (HPC) environment. We summarize the overall problem as follows: how can the same level of computational performance be achieved using less electrical power? Equivalently, how can computational performance be increased using the same amount of electrical power? In this dissertation we present performance and architecture models of DVFS as well as the Adagio runtime system. The performance model recasts the question as an optimization problem that we solve using linear programming, thus establishing a bound on potential energy savings. The architectural model provides a low-level explanation of how memory bus and CPU clock frequencies interact to determine execution time. Using insights provided from these models, we have designed and implemented the Adagio runtime system. This system realizes near-optimal energy savings on real-world scientific applications without the use of training runs or source code modification, and under the constraint that only negligible delay will be tolerated by the user. This work has opened up several new avenues of research, and we conclude by enumerating these.

No description available.

Cette thèse présente une contribution à la conception de circuits asynchrones Quasi Insensibles aux Délais (QDI) faible consommation. Une brève étude des méthodes d'estimation de l'énergie dans les circuits CMOS est présentée. Dans le deuxième chapitre, la méthodologie proposée sera présentée. Cette méthodologie utilise trois outils qui permettent la synthèse, l'optimisation et l'estimation d'énergie des circuits asynchrones QDI. La conception de ces circuits se fait à partir d'un langage de haut niveau (CHP). Le troisième chapitre expose une étude sur les choix d'architectures lors de la conception des circuits asynchrones QDI en utilisant la méthodologie proposée. Une comparaison avec les équivalents synchrones des architectures étudiées sera aussi montrée. Finalement, le quatrième chapitre présente une technique pour réduire la consommation d'un circuit en régulant la tension d'alimentation avec un asservissement à boucle fermée pour contrôler la tension d'alimentation.

Since the development of SoC is very fast, how to reduce the power consumption of SoC and improve the performance of SoC has become a very important issue. The power consumption of a system depends upon the hardware and software of a system. To overcome the issue of power consumption, the hardware circuit provides multi-voltage method to reduce task power consumption. On the other hand, the software tool decides the exact voltage for each task to minimize the total power consumption and finds a pipelined schedule of the periodic tasks to enhance the total throughput. In this thesis, a Tabu search is used to solve the voltage mapping and resource mapping problems of multi-voltage systems. This goal of this Tabu search is to find the solution with minimal power consumption for the multi-voltage system under the time constraints and resource constraints at the same time in the multi-voltage system to. Under the throughput constraints we use Tabu search to find solutions including the task¡¦s execution voltage and resource mapping, and then use list pipelined scheduling to schedule task and data communication and check their correctness. This method can reduce total power consumption. Experimental results show that our proposed algorithm can decide the resources mapping and pipeline in seconds, and it can reduce the power consumption efficiently.

Circuit designers perform optimization procedures targeting speed and power during the design of a circuit. Gate sizing can be applied to optimize for speed, while Dual-VT and Dynamic Voltage Scaling (DVS) can be applied to optimize for leakage and dynamic power, respectively. Both gate sizing and Dual-VT are design-time techniques, which are applied to the circuit at a fixed voltage. On the other hand, DVS is a run-time technique and implies that the circuit will be operating at a different voltage than that used during the optimization phase at design-time. After some analysis, the risk of non-critical paths becoming critical paths at run-time is detected under these circumstances. The following questions arise: 1) should we take DVS into account during the optimization phase? 2) Does DVS impose any restrictions while performing design-time circuit optimizations?. This thesis is a case study of applying DVS to a circuit that has been optimized for speed and power, and aims at answering the previous two questions. We used a 45-nm CMOS design kit and flow. Synthesis, placement and routing, and timing analysis were applied to the benchmark circuit ISCAS?85 c432. Logical Effort and Dual-VT algorithms were implemented and applied to the circuit to optimize for speed and leakage power, respectively. Optimizations were run for the circuit operating at different voltages. Finally, the impact of DVS on circuit optimization was studied based on HSPICE simulations sweeping the supply voltage for each optimization. The results showed that DVS had no impact on gate sizing optimizations, but it did on Dual-VT optimizations. It is shown that we should not optimize at an arbitrary voltage. Moreover, simulations showed that Dual-VT optimizations should be performed at the lowest voltage that DVS is intended to operate, otherwise non-critical paths will become critical paths at run-time.

Decorticated flax contains a significant amount of shive content, which limits applications of flax fibre. Separation of shives from the fibre is essential to improve the quality of flax fibre. Pneumatic method and a Sorter were implemented to meet the above objective. Terminal velocities of individual flax fibre and shive particles were investigated and their width, length, and mass were recorded. A sorting method was used for separation of short and long fibre for two grades of fibre: Grade 1 and Grade 2, with initial fibre purities of 51% and 15%, respectively. The ranges of terminal velocities for shive and fibre particles were 1.13 to 4.09 m/s and 0.51 to 1.07 m/s, respectively, which were significantly different. Fibre purity of approximately 80% for Grade 1 and 66% for Grade 2 were recorded from sorting, which were a significant improvement when compared to the initial purities. This study demonstrated the potential of the pneumatic and sorting methods for improving fibre quality. With the increase in resistant strains of microorganisms to antibiotics, researchers have started to explore plant parts to discover new antimicrobials. Since medieval times all portions of plants were used medicinally. Plant tissues, including stems, possess secondary metabolites (SMs), which have known antimicrobial properties. The purpose of this study was to investigate: the presence of antimicrobial compounds in stem extracts of canola, flax, hemp, and sweet clover; and study sorption-desorption behavior of their powdered stem material. GC-MS analysis of all extracts showed the presence of many SMs, including fatty acids, terpenoids, steroids, and sterols, etc. Many of the SMs found in the extracts have previously shown antimicrobial activity against a broad spectrum of organisms according to literature. Water sorption isotherms of stems showed a typical IUPAC Type II sigmoid curve similar to natural fibres. Equilibrium moisture content (EMC) of canola and sweet clover was significantly higher than flax and hemp at 95% RH, which were all higher than the fibre saturation point of wood (27%). The preliminary investigation via GC-MS showed promising results and water absorptivity results of stems can be used as the initial key property for many applications. / October 2015