This master thesis presents, examines, designs, implements, and evaluates with respect to energy efficiency a secure and robust VoIP system. This system utilizes a Session Initiation Protocol (SIP) infrastructure assisted by a cloud service, specifically focusing on small to medium sized enterprises (SME) and homes. This research focuses on using inexpensive, flexible, commodity embedded hardware (specifically a Linksys WRT54GL wireless router for the local site with a customized operating system, specifically DD-WRT). The idea is to reduce the local site's power consumption to very low levels by examining which functions can be done in a cloud service rather than at the local site. The thesis presents the design of a low-power IP telephony system for the local site and the cloud site. A number of different usage scenarios and desirable features are described. The methodology for conducting a set of experiments is defined to perform stress-testing and to evaluate the low- power IP telephony system's design. The experiments concern the overall power consumption of the local site under various configurations, the VPN link's call capacity, the QoS metrics for the VoIP calls, the session request delay (SRD) and the registration request delay (RRD). The results from these experiments show that there is a potential for significant power savings when using the proposed design for an IP telephony system. / Detta examensarbete presenterar, undersöker, utformar, implementerar, och försöker att utvärdera ett säkert och robust VoIP-system med energieffektivitet. Detta system använder en Session Initiation Protocol (SIP)-infrastruktur med hjälp av en molntjänst med särskild inriktning på, små, och medelstora företag (SME) och hemmanvändare. Denna forskning fokuserar att använda en prisvärt, billig, flexibel, med program inbyggd hårdvara (speciellt en Linksys WRT54GL trådlös router för den lokala platsen med ett anpassat operativsystem DD-WRT). Tanken är att minska energiförbrukningen på, den lokala platsen till mycket låga nivåer genom att undersöka vilka funktioner, som kan köras på, ett molntjnst snarare än på, den lokala platsen. Avhandlingen presenterar utformningen av ett IP-telefonisystem på, den lokala platsen med ett lågt strömbehov och på, molntjänsten. Ett antal olika användningsförhållanden och önskvärda egenskaper är beskrivna. Metodiken för att genomföra en rad experiment definieras för att utföra stresstester och för att utvärdera designen av IP-telefonisystem med ett lågt effektbehov. I försöken experimenteras den totala energiförbrukningen av den lokala platsen under olika konfigurationer, VPN-länkens samtalskapacitet, QoS-mätning för VoIP-samtal, Session Request Delay (SRD) och Registration Request Delay (RRD). Resultaten från dessa experiment visar att det finns en potential för betydande energibesparing när du använder den föreslagna designen för en IP-telefoni system.

Indoor ice rinks use 1091 MWh per annum for ice hockey based on statistics from over 100 Swedish ice rinks (Stoppsladd, 2011).The refrigeration system contributes 35 to75%( (Rogstam, 2010) of total energy usage in ice rinks with average value of 43% (Stoppsladd, 2010) for indoor to 75% for outdoor ice rinks. The basic aim of project is to reduce energy consumption in Swedish ice rinks and scope is for indoor and outdoor ice rinks in cold and mild summer climatic conditions like Sweden. To achieve target of energy reduction in ice rinks actual heat loads on outdoor bandy ice rink are being estimated along with performance analysis of refrigeration machine. The refrigeration system, heat loads on ice surface and their correlation is studied and analyzed in detail for Norrtälje Outdoor bandy ice rink for four warm days and whole season 2010-2011. The tricky and significant task of validation of input climate data for accurate heat loads calculations is completed with Swedish Metrological & Hydrological (SMHI) climate model data, correlations and related web based geographical data. The heat loads (conductive, convective and radiant) on outdoor bandy ice rink are calculated through thermodynamic relations with validated input climate data and measurements where as refrigeration system performance is monitored and analyzed with ClimaCheck(CC) instrumentation. The average cooling capacity is calculated for four warm days by CC internal method and actual cooling energy produced is obtained by practically assumed COP of system with aid of MYCOM compressor software. The cooling capacity and heat loads on ice surface are compared and analyzed considering energy usage affecting parameters and weather parameters like temperature, wind speed, relative humidity and solar load. The convection and condensation are contributing 75%, radiation 18%, ice resurfacing 4% and ground and header heat gain 3% to total heat loads on ice sheet for whole season. The deviation between total cooling energy produced by refrigeration machine and total heat load energy is found 19% and 27% for four warm days and whole season 2010-2011.The deviation is due to overestimation of heat losses from compressor’s body, compressor’s on and off operations, overestimated radiation heat load due to unmeasured negative radiation and lack of actual ice resurfacing heat load evaluation. The developed model in MS Excel allows comparison of field climate data with SMHI model data, indoor and outdoor ice rinks in terms of predicted energy usage by refrigeration system and in total and acts as decision tool to choose for building an indoor/outdoor ice rink.

With the increasing capacity of today's integrated circuits, a number ofheterogeneous  system-on-chip (SoC)  architectures  in embedded  systemshave been proposed. In order to achieve energy and design cost efficientstreaming applications  on these  systems, new design  space explorationframeworks  and  performance  analysis  approaches are  required.   Thisthesis  considers three state-of-the-art  SoCs architectures,  i.e., themulti-processor SoCs (MPSoCs)  with network-on-chip (NoC) communication,the hybrid CPU/FPGA architectures, and the run-time reconfigurable (RTR)FPGAs.  The main topic of the  author?s research is to model and capturethe  application  scheduling,  architecture  customization,  and  bufferdimensioning  problems, according to  the real-time  requirement.  Sincethese  problems  are NP-complete,  heuristic  algorithms and  constraintprogramming solver are used to compute a solution.For  NoC  communication  based  MPSoCs,  an  approach  to  optimize  thereal-time    streaming    applications    with   customized    processorvoltage-frequency levels and memory  sizes is presented. A multi-clockedsynchronous  model  of  computation   (MoC)  framework  is  proposed  inheterogeneous  timing analysis and  energy estimation.   Using heuristicsearching  (i.e., greedy  and  taboo search),  the  experiments show  anenergy reduction (up to 21%)  without any loss in application throughputcompared with an ad-hoc approach.On hybrid CPU/FPGA architectures,  the buffer minimization scheduling ofreal-time streaming  applications is addressed.  Based  on event models,the  problem  has  been  formalized  decoratively  as  constraint  basescheduling,  and  solved  by  public domain  constraint  solver  Gecode.Compared  with  traditional  PAPS  method,  the  proposed  method  needssignificantly smaller  buffers (2.4%  of PAPS in  the best  case), whilehigh throughput guarantees can still be achieved.Furthermore, a  novel compile-time analysis approach  based on iterativetiming  phases is  proposed  for run-time  reconfigurations in  adaptivereal-time   streaming   applications  on   RTR   FPGAs.   Finally,   thereconfigurations analysis and design trade-offs analysis capabilities ofthe proposed  framework have been  exemplified with experiments  on bothexample and industrial applications. / Andres

While global efforts are made to reduce the emission of greenhouse gases and move towards a more sustainable society, the global energy demand is continuing to increase. Building energy consumption represents 20-40% of the world's total energy use, and Heating, Ventilation, and Air Conditioning (HVAC) answer for around 50% of this amount. Only a small share of the European Union's building stock is considered to be energy efficient, and many of these buildings will continue to operate until the year 2050 and on-wards. The main objective of this thesis was to benchmark the economic and environmental implications of increasing building heating efficiency. To answer the framed research questions, an experimental study was carried out. In the study, a machine learning based solution was constructed and then implemented in a multi-tenant building for 24 days. Using an Artificial Neural Network a new heating curve was predicted, based on historical data from the building. The post-experimental data was then analyzed using STATA as statistical software tool. The results show that the new heating curve was able to reduce the heating system supply temperature by 1.9°C, with a decrease in average indoor temperature of 0.097°C. The decrease in supply temperature resulted in a reduction of energy expenditure by approximately 10%. Using the new building specific heating curve, yearly cost reductions of almost 11,700SEK could be achieved. Furthermore, the increased efficiency was able to reduce CO2 emissions by 127,5kg yearly. This results helps shed light on the general weaknesses in building heating systems out there today, and shows that there is great potential of reducing building energy consumption in cost effective ways. Although the implemented solution might not be generally applicable for all building owners out there, it should act as an eye opener for building owners and help motivate them into assessing their building operation and start looking into new technologies. Moreover, the study provides legible incentives for both building owners and the society to further work together towards a more efficient and sustainable society.

Large scale data centers (DC) and high performance computing (HPC) systems require more and more computing power at higher energy efficiency. They are already consuming megawatts of power, and a linear extrapolation of trends reveals that they may eventually lead to unrealistic power consumption scenarios in order to satisfy future requirements (e.g., Exascale computing). Conventional complementary metal oxide semiconductor (CMOS)-based electronic interconnects are not expected to keep up with the envisioned future board-to-board and chip-to-chip (within multi-chip-modules) interconnect requirements because of bandwidth-density and power-consumption limitations. However, low-power and high-speed optics-based interconnects are emerging as alternatives for DC and HPC communications; they offer unique opportunities for continued energy-efficiency and bandwidth-density improvements, although cost is a challenge at the shortest length scales. Plasmonics-based interconnects on the other hand, due to their extremely small size, offer another interesting solution for further scaling operational speed and energy efficiency. At the device-level, CMOS compatibility is also an important issue, since ultimately photonics or plasmonics will have to be co-integrated with electronics. In this paper, we survey the available literature and compare the aforementioned interconnect technologies, with respect to their suitability for high-speed and energy-efficient on-chip and offchip communications. This paper refers to relatively short links with potential applications in the following interconnect distance hierarchy: local group of racks, board to board, module to module, chip to chip, and on chip connections. We compare different interconnect device modules, including low-energy output devices (such as lasers, modulators, and LEDs), photodetectors, passive devices (i.e., waveguides and couplers) and electrical circuitry (such as laserdiode drivers, modulator drivers, transimpedance, and limiting amplifiers). We show that photonic technologies have the potential to meet the requirements for selected HPC and DC applications in a shorter term. We also present that plasmonic interconnect modules could offer ultra-compact active areas, leading to high integration bandwidth densities, and low device capacitances allowing for ultra-high bandwidth operation that would satisfy the application requirements further into the future.

Several studies have shown that the buildings envelope optical properties are important in terms of energy use and thermal comfort level. However, no study has been found in regard of the optimal optical properties for the building envelope. Moreover, developments in the coil-coating industry have made possible to design cost effective optical selective surfaces for the construction sector. Based on the above mentioned, this study pretends to map the envelope optimal optical properties minimizing the energy use for large-open-volume buildings locates in Stockholm, Copenhagen, Liverpool, Amsterdam, Berlin, Vienna, Bern, Rome, and Madrid.A building could be seen as a very complex solar energy conversion system, which is very difficult to describe accurately. Nonetheless, it is possible to use Building Energy Simulation (BES) tools to model, to some extent, its thermal performance under many simplistic assumptions. The simulation tool TRNSYS 17 and the optimization tool GenOpt were selected for this study. Additionally, detailed small-open-volume building thermal performance data, obtained during passive measurements from the steel manufacturer SSAB, in Borlänge-Sweden, were used to assess the methodology for the creation of the large-open-volume simulation models. The variations in large-open-volume building design around Europe are not well documented, which constitutes one of the major impediments for this research. However, detailed European historical building U-value data from the European Union project called iNSPiRe made it possible to achieve the objective of this study.The simulation work showed, that the building envelope optimal optical properties are related to the magnitude of the heating and cooling loads. Consequently, GenOpt was used to plot the sensitivity of the building envelope optimal optical properties to the ratio between the heating demand and the total energy demand (Qheat/Qtotal). In regard to the large-open-volume building optimal optical properties in the selected locations, it was found that the allocation of optimal optical properties does not lead to significant energy savings in locations with relatively low solar availability and high thermal insulation levels. Nonetheless, a final envelope optical properties study for a small-open-volume building model based on three existing buildings differing only on their optical properties was made for 243 world-capital cities. The simulations reinforced the results for the large-open-volume building in the European locations, and additionally showed huge energy savings potential for most of the world capital cities. This investigation restates the results obtained by Joudi (2015), “Possible energy savings by the smart choice of optical properties on the interior and exterior surfaces of the building.” / Varios estudios han demostrado que las propiedades ópticas de las envolturas delos edificios son importantes en términos de consumo energético y de nivel deconfort térmico. Sin embargo, no se ha encontrado estudio alguno con respecto alas propiedades ópticas óptimas para las envolturas de los edificios.Conjuntamente, los desarrollos en la industria de bobinas metálicas revestidas hanhecho posible diseñar superficies selectivas rentables para el sector de laconstrucción. Basándose en lo anteriormente expuesto, este estudio pretendemapear las propiedades ópticas óptimas de la envoltura que minimizan el uso deenergía para edificios de gran volumen abierto localizados en Estocolmo,Copenhague, Liverpool, Ámsterdam, Berlín, Viena, Berna, Roma y Madrid.Un edificio podría ser visto como un sistema de conversión de energía solar muycomplejo, que es muy difícil de describir con precisión. No obstante, es posibleutilizar las herramientas de Simulación de Energía de Edificios (BES) para modelar,hasta cierto punto, su rendimiento térmico bajo una considerable cantidad desuposiciones simplistas. El programa de simulación TRNSYS 17 y el programa deoptimización GenOpt fueron seleccionadas para este estudio. Adicionalmente, conel fin de evaluar la metodología utilizada para la creación de los modelos desimulación para edificios de gran volumen abierto, se utilizaron datos detallados derendimiento térmico de edificios de pequeño volumen abierto, obtenidos durantemediciones pasivas del fabricante de acero SSAB, en Borlänge-Suecia. Lasvariaciones en el diseño de edificios de gran volumen abierto en toda Europa noestán bien documentadas, lo que constituye uno de los principales impedimentospara esta investigación. Sin embargo, los datos detallados y en orden cronológicode los niveles de aislamiento térmico (U-value) en la construcción europea,recopilados por el proyecto de la Unión Europea llamado iNSPiRe, permitieronalcanzar el objetivo de este estudio.El trabajo de simulación demostró que las propiedades ópticas óptimas de laenvoltura del edificio están relacionadas con la magnitud de las cargas decalefacción y refrigeración. En consecuencia, GenOpt fue utilizado para graficar lasensibilidad de las propiedades ópticas óptimas de la envolvente del edificio conrespecto a la proporción entre la demanda de calefacción y la demanda total deenergía (Qheat/Qtotal). En cuanto a las propiedades ópticas óptimas del edificio degran volumen abierto en las ubicaciones seleccionadas, se encontró que laasignación de propiedades ópticas óptimas no conduce a ahorros de energíasignificativos en ubicaciones con disponibilidad solar relativamente baja y altosniveles de aislamiento térmico. Sin embargo, un último estudio de propiedadesópticas de envolvente para un modelo de construcción de pequeño volumen abiertobasado en tres edificios existentes que difieren sólo en sus propiedades ópticas serealizó para 243 capitales mundiales. Las simulaciones reforzaron los resultadospara el edificio de gran volumen abierto en las localidades europeas, y ademásmostraron un enorme potencial de ahorro de energía para la mayoría de lascapitales mundiales. Esta investigación reitera los resultados obtenidos por Joudi (2015), "Posibles ahorros de energía por la elección inteligente de propiedadesópticas en las superficies interiores y exteriores del edificio".

Syftet med examensarbetet är att ta fram ett förslag som är flexibelt, hållbart och kananvändas vid produktion av akutbostäder för hemlösa i Sverige. I vidare studier beaktasmaterial och ytteffektivitet av akutbostäder för strukturellt hemlösa i samhället.Resultatet visar på att en flexibel bostad med smarta servicelösningar kan användas tillmer än bara en akutbostad för hemlösa. Den kan även användas som hyresbostad,seniorboende eller annan typ av boende som behövs för tillfället. Genom intervjuer medfastighetsägare och myndighetspersoner har en kravlista sammanställts baserat på derasarbetserfarenheter.  Examensarbetet ger enbart ett förslag som kan användas som underlag vid uppförandet avakutbostäder eftersom inga egna beräkningar har behandlats. För att få ett tillförlitligtresultat bör arkitekter och andra involverade konsulter inom byggbranschen intervjuas. / The purpose of this project is to develop a flexible and sustainable proposal forproduction of emergency housing for homeless people in Sweden. This study will alsosuggest what should be considered in terms of materials and solutions in flexible,emergency housing for structurally homeless in society.The result shows that a flexible housing can be used for more than just emergencyhousing for the homeless. It could also be used as rental housing, senior housing or otherhousing needed. Through interviews with property owners and government officials, a listof requirements has been compiled based on their work experience.The thesis can be used as a foundation for further investigation in construction of emergency housing. Architects and other consultants involved in the building industryshould be interviewed to obtain a reliable result.

Au cours des dernières années, l’utilisation des téléphones portables et tablettes s’est vue croitre de façon significative. De plus, la disponibilité et l’ubiquité de l’accès sans fil a permis de créer un environnement dans lequel les utilisateurs partout où ils sont accèdent en tout temps à des services se trouvant dans le cloud. Cet environnement appelé cloud sans fil consomme une quantité d’énergie significative et est responsable d’émissions considérables en carbone. Cette consommation massive d’énergie et émissions en carbone deviennent un problème majeur dans le secteur des technologies de la communication. Dans ce contexte, nous nous intéressons dans cette thèse à la réduction de la consommation d’énergie et des empreintes en carbone des réseaux d’accès de type mesh et réseaux de campus ainsi que les data centers des infrastructures cloud. Dans la première partie, nous nous intéressons aux réseaux mesh de type TDMA. Nous proposons des solutions pour le problème de routage et ordonnancement des liens qui permettent de réduire la consommation d’énergie dans le réseau. Nous étendons par la suite cette approche pour les réseaux de campus dans un contexte compatible avec le paradigme SDN. Dans la deuxième partie, nous nous intéressons à la réduction de la consommation l’énergie et des empreintes en carbone des infrastructures cloud distribuées, avec des couts variables de l’électricité et d’émission en carbone. Nous proposons des approches d’optimisations dans deux cas de figures : le cas d’un fournisseur cloud souhaitant réduire ses couts et dans le cas où les utilisateurs cloud spécifient des contraintes en carbone sous forme de Green SLA. / Over the last decade, there has been an increasing use of personal wireless devices, such as laptops, smartphones and tablets. The widespread availability of wireless access created an environment in which anywhere at anytime users access data and services hosted in cloud infrastructures. However, such wireless cloud network consumes a non-negligible amount of energy and generates a considerable amount of carbon, which is becoming a major concern in IT industry. In this context, we address the problem of reducing energy consumption and carbon footprint, as well as building green infrastructures in the two different parts of the wireless cloud: (i) wireless access networks including wireless mesh and campus networks, and (ii) data centers in a cloud infrastructure. In the first part of the thesis, we present an energy-efficient framework for joint routing and link scheduling in multihop TDMA-based wireless networks. At a later stage, we extended this framework to cover campus networks using the emerging Software Defined Networking (SDN) paradigm. In the second part of this thesis, we address the problem of reducing energy consumption and carbon footprint of cloud infrastructures. Specifically, we propose optimization approaches for reducing the energy costs and carbon emissions of a cloud provider owning distributed infrastructures of data centers with variable electricity prices and carbon emissions in two different setups: the case of a cloud provider trying to reduce its carbon emissions and operational costs as well as the case where green constraints are specified by the cloud consumers in the form of Green SLAs.

L'amélioration de l'efficacité énergétique est devenue aujourd'hui une nécessité dans tous les domaines techniques. La réduction de la consommation, et donc du bilan carbone, est placée parmi les priorités mondiales tel que le paquet énergie-climat 2020 de l'Union Européenne.Les systèmes ferroviaires font partie des plus grands consommateurs d'énergie. Des solutions électriques sont développées pour réduire les pertes dans ces systèmes, optimiser la consommation et donc réduire la facture énergétique globale. Étant donné la diversité de ces systèmes, deux catégories principales sont considérées. La première regroupe les lignes urbaines caractérisées par une électrification en mode DC et un trafic relativement dense. Dans ce cas, l'énergie de freinage brûlée dans les rhéostats des trains constitue une perte considérable. La solution proposée consiste à récupérer cette énergie à l'aide d'un DC micro-grid installé dans une station passager. Elle permettra une interaction avec son environnement non-ferroviaire comme par exemple réutiliser cette énergie pour charger des bus électriques hybrides stationnant à proximité. Ce micro-grid contient un premier convertisseur DC/DC qui récupère l’excès d'énergie de freinage d'un train et l'injecte dans un DC busbar. Un deuxième convertisseur DC/DC va ensuite la stocker dans un système de stockage hybride pour que le bus électrique puisse se charger une fois branché au DC busbar. Le micro-grid est relié au réseau par un onduleur réversible AC/DC de faible puissance. L'ensemble est géré localement par un système gestion de puissance. Une évaluation énergétique montre que cette solution est intéressante lorsqu’un investissement, station de charge, est nécessaire pour charger les bus. En plus, dans le cas du DC micro-grid, aucun contrat avec le fournisseur d’électricité n’est nécessaire. La stabilité du système est aussi étudiée et une commande de stabilisation, le backstepping, est appliquée. Ce nouveau concept d’une future station intelligente permettra au système ferroviaire de communiquer avec son environnement qui est en pleine évolution.La deuxième catégorie est constituée par les lignes régionales et les lignes à grandes vitesses fonctionnant en mode AC. Contrairement au cas précédent, l’excès d’énergie de freinage est renvoyé à travers les sous-stations d’alimentation. Par conséquence, une deuxième solution propose la réduction de la consommation totale par l’optimisation du profile de vitesse de chaque train et la synchronisation de la grille horaire. Ceci est réalisé à l’aide d’un algorithme d’évolution différentielle. Chaque profile de vitesse est découpé en zones auxquelles sont attribuées des paramètres de conduite. L'optimisation de ces derniers permet de générer un nouveau profile de conduite optimal. Les résultats montrent la possibilité de faire des économies d’énergie tout en respectant la ponctualité des trains. / Increasing energy efficiency is nowadays a requirement in all technical fields. The reduction of global consumption, thus carbon footprint, has become the world's priority, as for example, the climate and energy package of the European Union.Railways' share of energy consumption is one of the highest. Electrical solutions are developed in order to reduce these systems' losses, optimize their consumption and reduce global energy bill. Given their diversity, two main categories are considered in this study. The first one consists of urban lines that are characterized by a DC electrification and a relatively dense traffic. In this case, braking energy burned in trains' rheostats represents the main share of losses. The proposed solution is to recuperate this energy using a DC micro-grid implemented in a passengers' station. It allows an interaction with the non-railway electrical environment, for example, re-using this energy in charging electric hybrid buses parked nearby. The excess of braking energy is recuperated using a DC/DC converter and injected into a DC busbar. A second DC/DC converter will store it in a hybrid storage system. It will then serve to charge the buses connected to the DC busbar. The micro-grid is also connected to the grid using a low power AC/DC converter. A power management system ensures optimizing power flow between different components. An energy evaluation showed that this solution is a good Investment especially because no contract is needed with the energy provider. The system's stability is studied and a stabilizing command, the backstepping, is applied. This new smart station allows railways to communicate, energetically, with its evolving environment.The second category is suburban and high speed lines that are AC electrified. Contrarily to the previous case, braking energy is reinjected to the upper grid through substations. Therefore, a second solution is to reduce global energy consumption by optimizing trains' speed profiles and timetable's synchronization. It is done using a differential evolution algorithm. Each speed profile is divided into zones to which are associated driving parameters. The optimization of the latter allowed generating new optimal speed profiles and a less-consuming timetable. Simulation results showed that it is possible to make important energy savings while respecting train's punctuality.

Ces dernières années, de nouvelles problématiques sont nées au vu des considérations écologiques de plus en plus présentes dans notre société. Dans le domaine de la technologie de l'Information, les centres de calcul consomment actuellement environ 1.5% de l'électricité mondiale. Cela ne cesse d’augmenter en raison de l'évolution de nombreux domaines et particulièrement du Cloud Computing. Outre cet aspect environnemental, le contrôle de la consommation d’énergie fait désormais partie intégrante des paramètres de Qualité de Service (QoS) incombant aux fournisseurs de services de Cloud Computing. En effet, ces fournisseurs de services à la demande proposent à leurs utilisateurs un contrat de QoS, appelé SLA (Service Level Agreement), qui définit de manière précise la qualité de service qu’ils s’engagent à respecter. Le niveau de QoS proposé influence directement la qualité d’utilisation des services par les utilisateurs, mais aussi la consommation et le rendement général de l’ensemble des ressources de calcul utilisées, impactant fortement les bénéfices des fournisseurs de services.Le Cloud Computing étant intrinsèquement lié à la virtualisation des ressources de calcul, une élaboration de modèles d’architecture matérielle et logicielle est proposée afin de définir les caractéristiques de l’environnement considéré. Ensuite, une modélisation détaillée de paramètres de QoS en termes de performance, de sûreté de fonctionnement, de sécurité des données et de coûts est proposée. Des métriques associées à ces paramètres sont définies afin d’étendre les possibilités d'évaluation des SLA. Ces modélisations constituent la première contribution de cette thèse.Il convient alors de démontrer comment l’utilisation et l’interprétation de plusieurs métriques de QoS ouvrent la possibilité d'une analyse plus complexe et plus fine de la perspicacité des algorithmes de placement. Cette approche multi-critères leur apporte des informations importantes sur l’état de leur système qu’ils peuvent analyser afin de gérer le niveau de chaque paramètre de QoS. Ainsi, quatre métriques antagonistes, incluant la consommation énergétique, ont été sélectionnées et utilisées conjointement dans plusieurs algorithmes de placement de manière à montrer leur pertinence, l’enrichissement qu’elles apportent à ces algorithmes, et comment un fournisseur de service peut tirer profit des résultats d’une optimisation multi-objectifs. Cette seconde contribution présente un algorithme génétique (GA) ainsi que deux algorithmes gloutons. L’analyse du comportement de l'algorithme génétique a permis de démontrer différents intérêts d’une optimisation multi-critères appliquée à des métriques de QoS habituellement ignorées dans les études dédiées au Cloud Computing.La troisième contribution de cette thèse propose une étude de l’impact de l'utilisation des métriques de QoS sur l’ordonnancement de machines virtuelles au cours du temps. Pour cela, le simulateur CloudSim a été exploité et étendu afin d'améliorer ses fonctionnalités de gestion de consommation énergétique. Tout d’abord par l’ajout du DVFS (Dynamic Voltage & Frequency Scaling) apportant une gestion dynamique très précise des fréquences de fonctionnement CPU, puis la possibilité de reconfiguration de machines virtuelles et enfin par la gestion dynamique des évènements. Les simulations effectuées mettent en jeu l'ensemble de ces outils énergétiques ainsi que les algorithmes de placement et évaluent chacune des métriques de QoS sélectionnées. Ces simulations donnent une vision temporelle de l’évolution de celles-ci, en fonction des algorithmes utilisés et de plusieurs configurations d’optimisation du GA. Cela permet d'analyser sous différents angles le comportement des algorithmes gloutons, l'impact des optimisations du GA, et l'influence des métriques les unes par rapport aux autres.Une collaboration a pu être établie avec le laboratoire CLOUDS Laborartory de Melbourne, dirigé par Prof. Rajkumar Buyya. / In recent years, new issues have arisen in environmental considerations, increasingly pointed out in our society. In the field of Information Technology, data centers currently consume about 1.5% of world electricity. This increasing is due to changes in many areas, especially in Cloud Computing. Besides this environmental aspect, the management of energy consumption has become an important field of Quality of Service (QoS), in the responsibility of Cloud providers. These providers propose a QoS contract called SLA (Service Level Agreement), which specify the level of QoS given to users. The level of QoS offered directly influences the quality of the users' utilization, but also the overall energy consumption and performance of computing resources, which strongly affect profits of the Cloud providers. Cloud computing is intrinsically linked to the virtualization of computing resources. A model of hardware and software architecture is proposed in order to define the characteristics of the environment considered. Then, a detailed modeling of QoS parameters in terms of performance, dependability, security and cost is proposed. Therefore, QoS metrics, associated to these parameters are defined in order to extend the possibilities for evaluating the SLA. These models represent the first contribution of this thesis. Then, it is necessary to illustrate how the use and interpretation of several QoS metrics open the possibility of a more complex and precise analysis of algorithms' insight. This multi-criteria approach, that provides useful informations about the system's status can be analyzed to manage the QoS parameters' level. Thus, four antagonists metrics, including energy consumption, are selected and used together in several scheduling algorithms which allow to show their relevance, the enrichment given to these algorithms, and how a Cloud provider can take advantage of the results of this kind of multi-objective optimization. The second contribution presents a genetic algorithm (GA) and two greedy algorithms. The analysis of the genetic algorithm behavior allows to show different interests of a multi-criteria optimization applied to QoS metrics, usually ignored in studies dedicated to Cloud Computing. The third contribution of this thesis proposes a study of the impact of the use of QoS metrics in virtual machines scheduling. The simulator CloudSim has been used and expanded to improve its energy-aware tools. The DVFS (Dynamic Voltage & Frequency Scaling), providing a highly accurate dynamic management of CPU frequencies, the virtual machines reconfiguration, and the dynamic management of events have been included. The simulations involve all of these energy tools and placement algorithms, and evaluate each selected QoS metrics. These simulations allow to see the evolution in time of these metrics, depending on the algorithms used and the behavior of the GA in different optimizations configurations. This allows to analyze from different angles the behavior of greedy algorithms, the impact of optimizations GA, and the influence of these metrics one against the others.