Dissertations / Theses on the topic 'Serveurs (informatique) – Économies d'énergie'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Serveurs (informatique) – Économies d'énergie.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Santi, Nina. "Prédiction des besoins pour la gestion de serveurs mobiles en périphérie." Electronic Thesis or Diss., Université de Lille (2022-....), 2023. http://www.theses.fr/2023ULILB050.
Full textMulti-access Edge computing is an emerging paradigm within the Internet of Things (IoT) that complements Cloud computing. This paradigm proposes the implementation of computing servers located close to users, reducing the pressure and costs of local network infrastructure. This proximity to users is giving rise to new use cases, such as the deployment of mobile servers mounted on drones or robots, offering a cheaper, more energy-efficient and flexible alternative to fixed infrastructures for one-off or exceptional events. However, this approach also raises new challenges for the deployment and allocation of resources in terms of time and space, which are often battery-dependent.In this thesis, we propose predictive tools and algorithms for making decisions about the allocation of fixed and mobile resources, in terms of both time and space, within dynamic environments. We provide rich and reproducible datasets that reflect the heterogeneity inherent in Internet of Things (IoT) applications, while exhibiting a high rate of contention and interference. To achieve this, we are using the FIT-IoT Lab, an open testbed dedicated to the IoT, and we are making all the code available in an open manner. In addition, we have developed a tool for generating IoT traces in an automated and reproducible way. We use these datasets to train machine learning algorithms based on regression techniques to evaluate their ability to predict the throughput of IoT applications. In a similar approach, we have also trained and analysed a neural network of the temporal transformer type to predict several Quality of Service (QoS) metrics. In order to take into account the mobility of resources, we are generating IoT traces integrating mobile access points embedded in TurtleBot robots. These traces, which incorporate mobility, are used to validate and test a federated learning framework based on parsimonious temporal transformers. Finally, we propose a decentralised algorithm for predicting human population density by region, based on the use of a particle filter. We test and validate this algorithm using the Webots simulator in the context of servers embedded in robots, and the ns-3 simulator for the network part
Nitu, Vlad-Tiberiu. "Improving energy efficiency of virtualized datacenters." Phd thesis, Toulouse, INPT, 2018. http://oatao.univ-toulouse.fr/23799/1/NITU_Vlad%20Tiberiu.pdf.
Full textAit, Aba Massinissa. "Optimisation de l'énergie et de la performance d'applications sur des micro-serveurs hétérogènes." Electronic Thesis or Diss., Sorbonne université, 2020. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2020SORUS106.pdf.
Full textRecent applications, both in industry and research often need massive calculations. They have different hardware requirements in terms of computing speed, which leads to very high energy consumption of hardware platforms. Heterogeneous computing platforms offer a good compromise with high computing power while preserving the energy consumed to run high-performance parallel applications. They are therefore nowadays an interesting computing resource. In order to exploit the advantages offered by heterogeneity in terms of performance, efficient and automatic management of computing resources is becoming increasingly important to execute parallel applications. These new architectures have thus given rise to new scheduling problems that allocate and sequence calculations on the different resources by optimizing one or more criteria. The objective of this thesis is to determine an efficient scheduling of a parallel application on a heterogeneous resource system in order to minimize the total execution time (makespan) of the application while respecting an energy constraint. Two classes of heterogeneous platforms have been considered in our work: fully heterogeneous architectures that combine several processing elements (CPUs, GPUs, FPGAs), and hybrid platforms limited to two types of processors (CPU + GPU for example). We propose several application scheduling strategies on both platforms with two execution models. Preliminary experiments with the proposed algorithms using different applications and platforms of different sizes have shown good results compared to existing methods in the literature
Borgetto, Damien. "Allocation et réallocation de services pour les économies d'énergie dans les clusters et les clouds." Toulouse 3, 2013. http://thesesups.ups-tlse.fr/2100/.
Full textCloud computing has become over the last years an important paradigm in the computing landscape. Its principle is to provide decentralized services and allows client to consume resources on a pay-as-you-go model. The increasing need for this type of service brings the service providers to increase the size of their infrastructures, to the extent that energy consumptions as well as operating costs are becoming important. Each cloud service provider has to provide for different types of requests. Infrastructure manager then have to host all the types of services together. That's why during this thesis, we tackled energy efficient resource management in the clouds. In order to do so, we first modeled and studied the initial service allocation problem, by computing approximated solutions given by heuristics, then comparing it to the optimal solution computed with a linear program solver. We then extended the model of resources to allow us to have a more global approach, by integrating the inherent heterogeneity of clusters and the cooling infrastructures. We then validated our model via simulation. Usually, the services must face different stages of workload, as well as utilization spikes. That's why we extended the model to include dynamicity of requests and resource usage, as well as the concept of powering on or off servers, or the cost of migrating a service from one host to another. We implemented a simulated cloud infrastructure, aiming at controlling the execution of the services as well as their placement. Thus, our approach enables the reduction of the global energy consumption of the infrastructure, and limits as much as possible degrading the performances
Pamba, Capo-Chichi Medetonhan Shambhalla Eugène William. "Conception d’une architecture hiérarchique de réseau de capteurs pour le stockage et la compression de données." Besançon, 2010. http://www.theses.fr/2010BESA2031.
Full textRecent advances in various aeras related to micro-electronics, computer science and wireless networks have resulted in the development of new research topics. Sensor networks are one of them. The particularity of this new research direction is the reduced performances of nodes in terms of computation, memory and energy. The purpose of this thesis is the definition of a new hierarchical architecture of sensor networks usable in different contexts by taking into account the sensors constraints and providing a high quality data such as multimedia to the end-users. We present our hierachical architecture with different nodes and the wireless technologies that connect them. Because of the high consumtpionof data transmission, we have developped two data compression algortithms in order to optimize the use of the channel by reducing data transmitted. We also present a solution for storing large amount of data on nodes by integrating the file system FAT16 under TinyOS-2. X
Chebira, Mahmoud Sabri. "Définition d'une stratégie de gestion locale d'un réseau sans fil à contraintes d'échéances strictes et économie d'énergie." Paris 12, 2006. https://athena.u-pec.fr/primo-explore/search?query=any,exact,990003939310204611&vid=upec.
Full textMany applications appear in different fields with the democratization of wireless networks. Most of them concern the activities of office and home automation. However, what about industrial applications? What does the market currently propose by taking into account temporal and quality constraints of connection linked to this domain? With this thesis, we propose implementations for industrial processes which request a wireless link subjected to strict temporal and energetic constraints. After a comparative study of various standards of wireless networks without thread available on the market, we opted for the latest standard 802. 15. 4 (alias ZigBee). It is designed with energy consumption saving abilities. These particularities present a real advantage for industrial automation control applications. Nevertheless, in our work, we propose few modifications within this standard at the MAC layer level which will allow a bigger flexibility regarding the management of the busy bandwidth and the guarantee of temporal deliverance of messages. This study ends with an analysis of the methods used for interconnection of network cells concerning an energetically-optimized routing technique to be developed at the level of the network layer
Gouvy, Nicolas. "Routage géographique dans les réseaux de capteurs et d’actionneurs." Thesis, Lille 1, 2013. http://www.theses.fr/2013LIL10185/document.
Full textThis thesis is about wireless multi-hop networks such as sensor/actuator networks and actuator networks. Those networks are composed of independent entities which have limited computing and memory capabilities and are battery powered. They communicate through the radio medium and do not require any static infrastructure. In order to relay messages between actuators up to the base station, we use what is called "routing protocols". My works rely on CoMNet, the first geographic routing protocol which aims to adapt the network topology to the routed traffic in order to save energy. Nevertheless, CoMNet does not consider the consequences of those relocations more than in a one-hop way. We proposed MobileR (Mobile Recursivity), which anticipates the routing in a multi-hop manner through computations over its one-hop neighbors. Hence it can select the “best” next forwarding node according to its knowledge. Another important topic is that events are likely to be detected by multiple sensors and all of them transmit message toward the destination. But those messages are likely to cross over an intersection node. This crossing provokes useless oscillation for it and premature node death. The PAMAL (PAth Merging ALgorithm) routing algorithm detects those routing path crossing and provokes a path merging upstream and uses a packet aggregation downstream. Finally, the Greedy Routing Recovery (GRR) protocol takes controlled mobility into account in order to increase delivery rate on topology with holes or obstacles. GRR includes a dedicated relocation pattern which will make it circumvent routing holes and create a routing path
Choy, Laurent. "Vers un paradigme de programmation orienté workflow pour la résolution de méthodes d'algèbre linéaire sur des plateformes de calcul global à faible consommation énergétique." Lille 1, 2007. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/2007/50376-2007-Choy.pdf.
Full textFaheem, Yasir. "Routage avec économie d'énergie dans les réseaux de capteurs sans fils." Paris 13, 2012. http://scbd-sto.univ-paris13.fr/secure/edgalilee_th_2013_faheem.pdf.
Full textLimited battery power is one of the major stringent factors in deploying Wireless Sensor Networks (WSNs), in spite of their numerous applications both on small scale as inWireless Body Area Networks (WBANs) and on large scale as in agricultural and habitat monitoring. Especially, stationary sink based data gathering protocols for large scaleWSNs have limited network lifetime, because relay nodes around the sink quickly deplete their battery power due to high traffic loads, making the rest of the network unreachable to the sink. On the other hand, sink mobility improves network lifetime by distributing relay nodes’ energy consumption. However, mobile sink now has to periodically update the network about its changing position. This control traffic is non-negligible for low power, limited capacity sensors as it induces energy consumption problem. In this thesis, we are considering energy efficient routing protocols in the context of WBANs and large scale WSNs. Moreover, we also address multi-channel assignment algorithm with the aim of minimizing power consumption and increasing network throughput. In the first part of this thesis, a deep analysis of the energy consumption of one hop vs multi-hop communications in WBANs is performed. In fact, recent advances in technology has led to the development of small, intelligent, wearable sensors which are capable of remotely performing critical health monitoring tasks, and then transmitting patient’s data back to health care centers over wireless medium. But to the day, energy also remains to be a big constraint in enhancing WBAN lifetime [Net12]. Some recent literature on WBANs proposes deliberate use of multi-hops to transfer data from a sensor to the gateway via relay health sensors as more energy efficient than single hop communication. There are studies which argue contrarily. In this context, we have analyzed the single vs multi-hop energy consumption effect for real very short range sensor devices. In the second part of this thesis, two distributed energy-efficient sink location update algorithms are proposed for large scale mobile sink WSNs. First algorithm, named SN- MPR, uses a combination of multi-point relay broadcast and a local path repair mechanism by means of which sink’s location update packets are forwarded only to nodes which are affected by sink mobility; the rest of the network does not receive these update messages. Next, a duty-cycle aware multi-point relay based algorithm which is a modified version of the SN-MPR algorithm is proposed. It allows non-relay nodes to switch-off their radios when communication is not desired. Simulation results show that the two aforementioned algorithms minimize network’s power consumption without compromising data delivery efficiency. The final part of this thesis deals with traffic-aware channel assignment problem in IEEE 802. 15. 4 standard-based heterogeneous WSNs which have rather high traffic rate requirements than low-rate scalar WSN applications. In fact, traditional single channel communication suffers from interferences caused by concurrent transmissions in the same neighborhood. These parallel transmissions waste battery power as multiple retransmis- sions are required before a packet can be successfully delivered at the destination due to frequent collisions. Moreover, already limited network throughput of the single channel communication protocols is further degraded at higher traffic rates due to increased colli-sions and congestion. On the other hand, concurrent transmissions over multiple channels not only reduce power consumption as packet collisions are minimized or eliminated depend- ing upon the efficiency of the concerned channel assignment algorithm, but also offer better network throughput and data delivery delays. Modern WSN platforms like crossbow’s Mi-caZ nodes [Mot12] are equipped with single, half-duplex IEEE 802. 15. 4 standard-based radio which can operate over sixteen multiple channels. In order to make effective use of multiple channels, a number of channel assignment algorithms have been proposed recently for WSNs. However, they are suitable for rather low-rate homogeneous WSNs, and they consider fixed physical channel widths. These multi-channel assignments increase network throughput, but they may not be able to ensure QoS requirements of high bandwidth de- manding multimedia traffic, as in the case of heterogeneous WSNs. In order to address the energy issue and at the same time increase network capacity, we propose a distributive Traffic-Aware Bandwidth-Adaptive (TABA) channel selection algorithm which enables the nodes to not only choose interference free channels in the neighborhood, but also to adapt channel-width to increase/decrease throughput according to varying traffic conditions
Soto, Lima María Consuelo. "Optimization methods for the memory allocation problems in embedded systems." Lorient, 2011. http://www.theses.fr/2011LORIS238.
Full textMemory allocation in embedded systems is one of the main challenges that electronic designers have to face. This part, rather difficult to handle is often left to the compiler with which automatic rules are applied. Nevertheless, a carefully tailored allocation of data to memory banks may lead to great savings in terms of running time and energy consumption. This thesis addresses various versions of the memory allocation problem. At each version the problem's complexity increases, i. E. , the number of constrains increases. The number of memory banks, bank capacities, sizes and number of accesses of data structures, and the conflicting data structures at each time interval are the main constrains handled in the memory allocationproblems. In this work we present an ILP formulation and some metaheuristics implemented for each problem version. We also assess ourmetaheuristics with the exact methods and other literature metaheuristics with the aim of highlighting what makes the success of metaheuristics for these problems
Kacem, Fadi. "Algorithmes exacts et approchés pour des problèmes d'ordonnancement et de placement." Thesis, Evry-Val d'Essonne, 2012. http://www.theses.fr/2012EVRY0007/document.
Full textIn this thesis, we focus on solving some combinatorial optimization problems that we have chosen to study in two parts. Firstly, we study optimization problems issued from scheduling a set of tasks on computing machines where we seek to minimize the total energy consumed by these machines while maintaining acceptable quality of service. In a second step, we discuss two optimization problems, namely a classical scheduling problem in architecture of parallel machines with communication delays, and a problem of placing data in graphs that represent peer-to-peer networks and the goal is to minimize the total cost of data access
Bessaoud, Karim. "Algorithmes auto-stabilisants pour les réseaux ad hoc." Versailles-St Quentin en Yvelines, 2013. http://www.theses.fr/2013VERS0048.
Full textIn this thesis, we propose three self-stabilizing algorithms for ad hoc wireless networks. The first one is an algorithm that builds a low weight connected dominating set, called backbone. The backbone is used to create a logical infrastructure in an ad hoc network. We show by simulation the efficiency of this algorithm in different contexts according to the semantics given to the weight of the nodes : the backbone may contain for instance the least mobile nodes or nodes with the highest battery level. The two other algorithms deal with energy conservation in wireless sensor networks. We propose two solutions based on topology control by reducing transmission powers, each one dedicated to a type of communication used by the sensors : communication between any pair of sensors or by diffusion. The proposed algorithms are formally proven and evaluated by simulation
Hamzaoui, Khalil Ibrahim. "Contribution à la modélisation de la consommation d'énergie dans un dispositif mobile." Thesis, Lille 1, 2018. http://www.theses.fr/2018LIL1I030/document.
Full textThe main goal of this thesis is to model the power consumption of a particular application running on a mobile device. We propose a model of energy behavior monitoring, we also describe a methodology to identify the parameters of the model. To this end, we analyzed a collection of experimental data collected during my tour de France in an electric wheelchair. We applied statistical tools to obtain the parameters of the model. Finally, we validate the model by comparing the results with other experimental data.The first case study compares the evolution of the energy cost in the mobile environments of the different components of smartphones based on several energy models.- The second case study deals with the evaluation, the measurements of the energy cost consumed and the problems encountered in the methods used for the evaluation of energy consumption. For a better evaluation, the case study of energy behavior was introduced using the virtual machines.- The third case study is based on the treatment of the results of the measurements obtained during my tour of France in a connected electric wheelchair. The goal is to anticipate resource management, realizing measurements, and then tracking energy behavior in a real and diverse environment. The model can be used to define an optimal frequency in terms of energy consumption for specific situations without degrading the quality of service desired by the user
Chen, Langshi. "Méthode de Krylov itératives avec communication et efficacité énergétique optimisées sur machine hétérogène." Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10114/document.
Full textIterative methods are frequently used in extremely large scale linear problems, such solving linear systems or finding eigenvalue/eigenvectors of matrices. As these iterative methods require a substantial computational workload, they are normally deployed on large clusters of distributed memory architectures communicated via MPI. When the problem size scales up, the communication becomes a major bottleneck of reaching a higher scalability because of two reasons: 1) Many of the iterative methods rely on BLAS-2 low level matrix vector kernels that are communication intensive. 2) Data movement (memory access, MPI communication) is much slower than processor's speed. In case of sparse matrix operations such as Sparse Matrix Vector Multiplication (SpMV), the communication even replaces the computation as the dominant time cost. Furthermore, the advent of accelerators/coprocessors like Nvidia's GPU make computation cost more cheaper, while the communication cost remains high in such CPU-coprocessor heterogeneous systems. Thus, the first part of our work focus on the optimization of communication cost of iterative methods on heterogeneous clusters. Besides the communication cost, power wall becomes another bottleneck of future exascale computing in recent time. Researches indicate that a power-aware algorithmic implementation strategy could efficiently reduce the power dissipation of large clusters. We also explore the potential energy saving implementation of iterative methods in our experimentation. Finally, both the communication optimization and energy efficiency implementation would be integrated into a GMRES method, which demands an auto-tuning framework to maximize its performance
Milent, Etienne. "Contribution à l'étude d'un actionneur asynchrone à contrôle vectoriel et de ses possibilités d'utilisation dans des applications embarquées." Compiègne, 1992. http://www.theses.fr/1992COMPD508.
Full textVillebonnet, Violaine. "Scheduling and Dynamic Provisioning for Energy Proportional Heterogeneous Infrastructures." Thesis, Lyon, 2016. http://www.theses.fr/2016LYSEN057/document.
Full textThe increasing number of data centers raises serious concerns regarding their energy consumption. These infrastructures are often over-provisioned and contain servers that are not fully utilized. The problem is that inactive servers can consume as high as 50% of their peak power consumption.This thesis proposes a novel approach for building data centers so that their energy consumption is proportional to the actual load. We propose an original infrastructure named BML for "Big, Medium, Little", composed of heterogeneous computing resources : from low power processors to classical servers. The idea is to take advantage of their different characteristics in terms of energy consumption, performance, and switch on reactivity to adjust the composition of the infrastructure according to the load evolutions. We define a generic methodology to compute the most energy proportional combinations of machines based on hardware profiling data.We focus on web applications whose load varies over time and design a scheduler that dynamically reconfigures the infrastructure, with application migrations and machines switch on and off, to minimize the infrastructure energy consumption according to the current application requirements.We have developed two different dynamic provisioning algorithms which take into account the time and energy overheads of the different reconfiguration actions in the decision process. We demonstrate through simulations based on experimentally acquired hardware profiles that we achieve important energy savings compared to classical data center infrastructures and management
Ingelrest, François. "Protocoles localisés de diffusion et économie d'énergie dans les réseaux ad hoc et de capteurs." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2006. http://tel.archives-ouvertes.fr/tel-00113869.
Full textLes réseaux de capteurs sont similaires aux réseaux ad hoc, car ils sont également décentralisés et autonomes. Un capteur est un petit appareil capable de surveiller son environnement. Des cas typiques d'utilisation peuvent être la surveillance de zones militaires (détection de mouvements) ou de forêts (détection d'incendie).
Parmi les problèmes communs à ces deux types de réseaux se trouve la diffusion. Dans une telle communication, un message est envoyé depuis un objet donné vers tous les autres du réseau. Les applications de ce processus sont nombreuses : découverte de routes, synchronisation... Comme les objets mobiles utilisent une batterie, il est nécessaire que la diffusion soit la plus économe possible d'un point de vue énergétique. Cela est généralement obtenu en réduisant la quantité de relais nécessaires, ou en limitant la puissance d'émission à chaque relais.
Le but de mon travail était d'étudier la diffusion dans les réseaux ad hoc et de capteurs, afin de mettre en lumière les caractéristiques et les défauts des mécanismes existants, puis d'en proposer de nouveaux, plus efficaces. Dans tous ces travaux, nous avons toujours voulu rester dans le domaine des solutions 'réalistes' : beaucoup des précédentes études utilisaient en effet des mécanismes centralisés, où une connaissance globale du réseau est nécessaire pour effectuer la diffusion. Nous nous sommes concentrés sur des solutions fiables et localisés, c'est-à-dire n'utilisant que des informations sur le voisinage de chaque noeud. Ce type de mécanisme permet également un passage à l'échelle simplifié, car la quantité d'informations nécessaire ne varie pas avec la taille du réseau. Nos études montrent de plus que ces solutions peuvent être aussi efficaces que les méthodes centralisées.
Puisque l'ajustement de portée est un mécanisme très important dans la conservation de l'énergie, nous avons proposé une méthode de diffusion originale, basée sur le concept de portée optimale de communication. Cette dernière est calculée de manière théorique grâce au modèle énergétique considéré, et représente le meilleur compromis entre l'énergie dépensée à chaque noeud et le nombre de relais nécessaires. Nous avons ainsi proposé deux protocoles différents basés sur ce concept, chacun étant plus spécifiquement adapté soit aux réseaux ad hoc (TR-LBOP), soit aux réseaux de capteurs (TR-DS).
Afin de réduire encore plus la consommation énergétique, nous avons étudié le fameux protocole centralisé nommé BIP. Son efficacité est due au fait qu'il considère la couverture obtenue par une seule émission omnidirectionnelle, au lieu de considérer chaque lien séparément. Nous avons proposé une solution localisée basée sur BIP, afin de construire incrémentalement une structure de diffusion. Nous avons montré de manière expérimentale que les résultats ainsi obtenus sont très proches de ceux fournis par BIP, notamment dans les réseaux de forte densité, tout en n'utilisant que des informations locales à chaque noeud.
Nous avons finalement considéré la suppression d'une hypothèse forte, largement répandue dans la communauté des réseaux ad hoc et de capteurs : l'utilisation d'un graphe du disque unitaire. Ce dernier définit la zone de communication d'un noeud comme étant un cercle parfait. Nous avons remplacé cette hypothèse par une autre plus réaliste afin d'en étudier les conséquences sur un protocole connu, le protocole de diffusion par relais multipoints (MPR). Nous avons montré que ce dernier ne fournit plus de résultats suffisants dans un tel environnement. Nous avons également proposé quelques modifications afin d'obtenir à nouveau de bons résultats.
Messai, Sarra. "Gestion de la mobilité dans les réseaux de capteurs sans fil." Thesis, Lyon, 2019. http://www.theses.fr/2019LYSE1183.
Full textWireless Sensor Networks (WSNs) are increasingly invading our lives. With the rise of Internet of Things (IoT), WSNs are used in applications that require observation of the physical world and data collection. However, many obstacles inherent to the specificities of WSNs must be overcome before reaching the maturity of this technology. Among these obstacles, the resource limitations such as energy, computing capability, bandwidth and storage capability of sensor nodes. In this thesis, we focus on mobility management as a solution to improve network performance in terms of energy consumption and optimization of data collection. A mobile wireless sensor network is a network in which at least the base station is mobile.We first looked at the case where only the base station is mobile. In this context, we proposed a network organization that leverages base station mobility to optimize data collection while reducing the dissipated energy by sensor nodes. The proposed organization is based on a grid architecture and an optimized base station mobility algorithm for collecting data. We implemented our solution in the NS-2 simulation environment. The obtained simulation results show clearly the improvement of our proposal compared to other existing approaches. Then we looked at the case where the mobility is total, in other words, each sensor node in the network can be mobile. In this case, we worked on the issue of key management to ensure the security of data collection. In this context, we proposed a new key management scheme based on the random key pre-distribution. Our solution has the particularity of ensuring the self-healing of the network where sensor nodes are compromised. We evaluated and implemented our solution and compared it with two other reference schemes to show its effectiveness
Horrein, Ludovic. "Gestion d’énergie décomposée d’un véhicule hybride intégrant les aspects thermiques via la représentation énergétique macroscopique." Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10078/document.
Full textHybrid vehicle uses several energy sources to reduce its fuel consumption. Energy management strategy (EMS) is one key element that enables to decide when and how to use these sources. For a single vehicle, different EMSs are possible. More the EMS considers constraints, more it is efficient but more it is difficult to solve in real time. The cabin heating system management who has a significant impact on the consumption is considered in a different EMS than the traction system management. Optimal EMS enables to evaluate the ideal power splitting to reach the minimal consumption. If this EMS can be theoretically evaluated in simulation, it cannot be implemented in real time. Nevertheless, it is possible to deduce a real time EMS using the results of the optimal EMS. This deduction is complex and no methodology is available to structure this conception. The objective of this thesis is to propose an EMS of a hybrid vehicle who includes the management of the traction system and of the heating system. To do that, the thermal exchanges are modeled and described using EMR (Energetic Macroscopic Representation) to unify all models. Experimental validations enable to evaluate the models accuracy. To finish, a step by step methodology is proposed to organize the development of a real time EMS. The results obtained with this real time EMS are closed to the optimal theoretical result
Wang, Yewan. "Évaluation et modélisation de l’impact énergétique des centres de donnée en fonction de l’architecture matérielle/ logicielle et de l’environnement associé." Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2020. http://www.theses.fr/2020IMTA0175.
Full textFor years, the energy consumption of the data center has dramatically increased followed by the explosion of demand in cloud computing. This thesis addresses the scientific challenge of energy modeling of a data center, based on the most important variables. With such modeling, an data center operator will be able to better reallocate / design the current / future data centers. In order to identify the energy impacts of hardware and software used in computer systems. In the first part of the thesis, to identify and characterize the uncertainties of energy consumption introduced by external elements: thermal effects, difference between identical processors caused by imperfect manufacturing process, precision problems resulting from power measurement tool, etc. We have completed this scientific study by developing a global power modeling for a given physical cluster, this cluster is composed by 48 identical servers and equipped with a direct expansion cooling system, conventionally used today for modern data centers. The modeling makes it possible to estimate the overall energy consumption of the cluster based on operational configurations and data relating to IT activity, such as ambient temperature, cooling system configurations and server load
Bechkit, Walid. "Security and energy saving in wireless sensor networks." Compiègne, 2012. http://www.theses.fr/2012COMP2045.
Full textWireless sensor networks (WSN) arc set to become one of the technologies which invade our everyday life. These networks suffer from several constraints mainly related to the resource limitations (energy, memory, etc. ) and to the harsh deployment environment. These limitations, coupled with the required security levels and network autonomy, create a clash between two key design issues : security and efficiency. Ln this thesis, we tackle these two design objectives and propose secure and energy saving solutions for smooth operation of WSN. We focus in the first part on key management which lays the foundation of security in WSN. We develop a new scalable key management scheme which makes use of unital design theory. Our scheme provides a good secure coverage of large scale networks with a low key storage overhead. We also address the problem of network resiliency through a generic class of hash-chain based key management schemes. Ln the second part, we address the energy saving challenging issue. We model the impact of the temperature on the radio communications in WSN and we propose new fully distributed temperature-aware connectivity-driven algorithms for energy saving in WSN. Furthermore, we address the energy aware routing and we propose a new weighted shortest path tree for convergecast traffic routing in WSN. We finally present AgroSens project funded by the ERDF and the Picardy regional council. Ln this project, we design and develop a WSN system for agriculture in Picardy. We present in this thesis the main project phases and we discuss the implementation of our architecture and communication protocols. We also present the main features of the testbed that we set up
Wang, Yuqi. "Suivi de l’état de santé des réseaux de distribution de chaleur." Thesis, Lille, 2019. http://www.theses.fr/2019LIL1I016.
Full textCompanies managing district heat networks, provide energy services to customers. Faced with rising energy costs and increasingly stringent regulatory and societal pressures, these energy service companies are seeking to control the system energy efficiency, aiming at improving their competitiveness and social image. In this context, the industrial needs to monitor the energy efficiency of the energy distribution system and its health state, in order to perform maintenance operations in case of problems. Due to the large amount of data and the desired short reaction time, it is necessary to develop methods to support the operator in the health state monitoring and in the maintenance decision-making process. These methods aim at providing appropriate health indicators of system's components, whose evolutions are easy to be interpreted by operators. Behavioral models of heat networks are used on the one hand to generate these indicators, and on the other hand to establish the links between component faults and their impacts on energy efficiency of the network. These links are analyzed to search for monitorable subsystems, i.e. to verify whether the faults to be monitored are structurally detectable, and to guide the indicators generation. Static models are retained for their genericity, simplicity of development, of calculation and of interpretation. Indicators generated using static models must be calculated when the system is operating in steady state. For this reason, we propose a method to determine the steady state time-windows. The computed indicators are then analyzed to give to the operator an information on the system's health state. Structural analysis shows that an indicator can be sensitive to different faults, and that a fault can influence the values of several indicators. The evolutions of these indicators can therefore be correlated. In order to use the information of the correlation between the indicators to isolate the fault,indicators values are analyzed in the space of indicators. We propose a dynamic clustering method to help the operator to monitor the health state of the network.The proposed approach is applied on a heat network plant managed by Veolia, the industrial partner of this thesis, using the data recorded during two months. The results show the efficiency of the proposed tools and methods
Tayeb, Jamel. "Optimisation des performances et de la consommation de puissance électrique pour architecture Intel Itanium/EPIC." Valenciennes, 2008. http://ged.univ-valenciennes.fr/nuxeo/site/esupversions/9eed6aef-dfaf-4a17-883f-d217a1d9a000.
Full textThis thesis proposes, in its first part, to extend the EPIC architecture of the Itanium processor family by providing a hardware stack. The principal idea defended here is that it is possible to close the existing performance gap between generic architectures and application specific designs to run virtual machines (FORTH,. NET, Java, etc). With this intention, we propose to reallocate dynamically a subset of the EPIC architecture’s resources to implement a hardware evaluation stack. Two implementations are proposed, both non-intrusive and compatible with existing binary codes. The fundamental difference between these stacks lies in their manager: software or hardware. The hardware controlled evaluation stack offers support for advanced functions such as the support of strongly typed evaluation stacks required by. NET’s CIL. Thus, we propose a single pass CIL binary translator into EPIC binary, using the hardware evaluation stack. In the second part of this thesis, we studied the energy efficiency of software applications. First, we defined a methodology and developed tools to measure the energy consumption and the useful work provided by the software. In a second time, we engaged the study of source code transformation rules in order to reduce/control the quantity of consumed energy by the software
Ghribi, Chaima. "Energy efficient resource allocation in cloud computing environments." Electronic Thesis or Diss., Evry, Institut national des télécommunications, 2014. http://www.theses.fr/2014TELE0035.
Full textCloud computing has rapidly emerged as a successful paradigm for providing IT infrastructure, resources and services on a pay-per-use basis over the past few years. As, the wider adoption of Cloud and virtualization technologies has led to the establishment of large scale data centers that consume excessive energy and have significant carbon footprints, energy efficiency is becoming increasingly important for data centers and Cloud. Today data centers energy consumption represents 3 percent of all global electricity production and is estimated to further rise in the future. This thesis presents new models and algorithms for energy efficient resource allocation in Cloud data centers. The first goal of this work is to propose, develop and evaluate optimization algorithms of resource allocation for traditional Infrastructutre as a Service (IaaS) architectures. The approach is Virtual Machine (VM) based and enables on-demand and dynamic resource scheduling while reducing power consumption of the data center. This initial objective is extended to deal with the new trends in Cloud services through a new model and optimization algorithms of energy efficient resource allocation for hybrid IaaS-PaaS Cloud providers. The solution is generic enough to support different type of virtualization technologies, enables both on-demand and advanced resource provisioning to deal with dynamic resource scheduling and fill the gap between IaaS and PaaS services and create a single continuum of services for Cloud users. Consequently, in the thesis, we first present a survey of the state of the art on energy efficient resource allocation in cloud environments. Next, we propose a bin packing based approach for energy efficient resource allocation for classical IaaS. We formulate the problem of energy efficient resource allocation as a bin-packing model and propose an exact energy aware algorithm based on integer linear program (ILP) for initial resource allocation. To deal with dynamic resource consolidation, an exact ILP algorithm for dynamic VM reallocation is also proposed. This algorithm is based on VM migration and aims at constantly optimizing energy efficiency at service departures. A heuristic method based on the best-fit algorithm has also been adapted to the problem. Finally, we present a graph-coloring based approach for energy efficient resource allocation in the hybrid IaaS-PaaS providers context. This approach relies on a new graph coloring based model that supports both VM and container virtualization and provides on-demand as well as advanced resource reservation. We propose and develop an exact Pre-coloring algorithm for initial/static resource allocation while maximizing energy efficiency. A heuristic Pre-coloring algorithm for initial resource allocation is also proposed to scale with problem size. To adapt reservations over time and improve further energy efficiency, we introduce two heuristic Re-coloring algorithms for dynamic resource reallocation. Our solutions are generic, robust and flexible and the experimental evaluation shows that both proposed approaches lead to significant energy savings while meeting the users' requirements
Huin, Nicolas. "Réseaux pilotés par logiciels efficaces en énergie." Thesis, Université Côte d'Azur (ComUE), 2017. http://www.theses.fr/2017AZUR4080/document.
Full textIn the recent years, the growth of the architecture of telecommunication networks has been quickly increasing to keep up with a booming traffic. Moreover, the energy consumption of these infrastructures is becoming a growing issue, both for its economic and ecological impact. Multiple approaches were proposed to reduce the networks' power consumption such as decreasing the number of active elements. Indeed, networks are designed to handle high traffic, e.g., during the day, but are over-provisioned during the night. In this thesis, we focus on disabling links and routers inside the network while keeping a valid routing. This approach is known as Energy Aware Routing (EAR). However current networks are not adapted to support the deployment of network-wide green policies due to their distributed management and the black-box nature of current network devices. The SDN and NFV paradigms bear the promise of bringing green policies to reality. The first one decouples the control and data plane and thus enable a centralized control of the network. The second one proposes to decouple the software and hardware of network functions and allows more flexibility in the creation and management of network services. In this thesis, we focus on the challenges brought by these two paradigms for the deployment of EAR policies. We dedicated the first two parts to the SDN paradigm. We first study the forwarding table size constraints due to an Increased complexity of rules. We then study the progressive deployment of SDN devices alongside legacy ones. We focus our attention on the NFV paradigm in the last part, and more particularly, we study the Service Function Chaining problem
Haderer, Nicolas. "APISENSE® : une plate-forme répartie pour la conception, le déploiement et l’exécution de campagnes de collecte de données sur des terminaux intelligents." Thesis, Lille 1, 2014. http://www.theses.fr/2014LIL10118/document.
Full textMobile crowdsensing is a new form of data collection that takes advantage of millions smart devices already deployed throughout the world to collect massively environmental or behavioral data from a population. Recently, this type of data collection has attracted interest from a large number of industrials and academic players in many areas, such as the study of urban mobility, environmental monitoring, health or the study of sociocultural attitudes. However, mobile crowdsensing is in its early stages of development, and many challenges remain to be addressed to take full advantage of its potential. These challenges include privacy, limited energy resources of devices, development of reward and recruitment models to select appropriates mobile users and dealing with heterogeneity of mobile platforms available. In this thesis, we aim to reconsider the architectural design of current mobile crowdsensing systems to provide a simple and effective way to design, deploy and manage data collection campaigns.The main contributions of this thesis are organize around APISENSE, the resulting platform of this research. APISENSE has been used to carry out a data collection campaign deployed over hundred of users in a sociological study and evaluated through experiments demonstrating the validity, effectiveness and scalability of our solution
Lorandel, Jordane. "Etude de la consommation énergétique de systèmes de communications numériques sans fil implantés sur cible FPGA." Thesis, Rennes, INSA, 2015. http://www.theses.fr/2015ISAR0036/document.
Full textWireless communication systems are still evolving since the last decades, driven by the growing demand of the electronic market for energy efficient and high performance devices. Thereby, new design constraints have appeared that aim at taking into account power consumption in order to improve battery-life of circuits. Current wireless communication systems commonly dissipate a lot of power. On the other hand, the complexity of such systems keeps on increasing through the generations to always satisfy more users at a high degree of performance. In this highly constrained context, FPGA devices seem to be an attractive technology, able to support complex systems thanks to their important number of resources. According to the FPGA nature, designers need to estimate the power consumption and the performance of their wireless communication systems as soon as possible in the design flow. In this way, they will be able to perform efficient design space exploration and make decisive implementation and optimization choices. Throughout this thesis, a power estimation methodology for hardware-focused FPGA device is described and aims at making design space exploration a lot easier, providing early and fast power and performance estimation at high-level. It also proposes an efficient way to efficiently compare several systems. The methodology is effective through an lP characterisation step and the development of their SystemC models. Then, a high level description of the entire system is realized from the SystemC models that have been previously developed. High-level simulations enable to check the functionality and evaluate the power and performance of the system. One of the contributions consists in monitoring the JP time-activities during the simulation. We show that this has an important impact on both power and performances. The effectiveness of the methodology has been demonstrated throughout several baseband processing chains of the wireless communication domain such as a SISO-OFDM generic chain, LTE transmitters etc. To conclude, the main limitations of the proposed methodology have been investigated and addressed
Gbaguidi, Fréjus A. Roméo. "Approche prédictive de l'efficacité énergétique dans les Clouds Datacenters." Electronic Thesis or Diss., Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1163.
Full textWith the democratization of digital technologies, the construction of a globalized cyberspace insidiously transforms our lifestyle. Connect more than 4 billion people at high speed, requires the invention of new concept of service provision and trafic management that are capable to face the challenges. For that purpose, Cloud Computing have been set up to enable Datacenters to provide part or total IT components needed by companies for timely services delivering with performance that meets the requirements of their clients. Consequently, the proliferation of Datacenters around the world has brought to light the worrying question about the amount of energy needed for their function and the resulting difficulty for the humanity, whose current reserves are not extensible indefinitely. It was therefore necessary to develop techniques that reduce the power consumption of Datacenters by minimizing the energy losses orchestrated on servers where each wasted watt results in a chain effect on a substantial increase in the overall bill of Datacenters. Our work consisted first in making a review of the literature on the subject and then testing the ability of some prediction tools to improve the anticipation of the risks of energy loss caused by the misallocation of virtual equipment on servers. This study focused particularly on the ARMA tools and neural networks which in the literature have produced interesting results in related fields. After this step, it appeared to us that ARMA tools, although having less performance than neural networks in our context, runs faster and are best suited to be implemented in cloud computing environments. Thus, we used the results of this method to improve the decision-making process, notably for the proactive re-allocation of virtual equipment before it leads to under-consumption of resources on physical servers or over-consumption inducing breaches of SLAs. Based on our simulations, this approach enabled us to reduce energy consumption on a firm of 800 servers over a period of one day by more than 5Kwh. This gain could be significant when considering the enormous size of modern data centers and projected over a relatively long period of time. It would be even more interesting to deepen this research in order to generalize the integration of this predictive approach into existing techniques in order to significantly optimize the energy consumption within Datacenters while preserving performance and quality of service which are key requirements in the concept of Cloud Computing
Dogeas, Konstantinos. "Energy Minimization, Data Movement and Uncertainty : Models and Algorithms." Electronic Thesis or Diss., Sorbonne université, 2022. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2022SORUS070.pdf.
Full textHigh performance computers (HPCs) is the go-to solution for running computationally demanding applications. As the limit of energy consumption is already achieved, the need for more energy efficient algorithms is critical.Taking advantage of the core characteristics of an HPC, such as its network topology and the heterogeneity of the machines, could lead to better scheduling algorithms. In addition, designing more realistic models, that grasp the features of real-life applications, is a work in the same direction of achieving better performance. Allowing scheduling algorithms to decide either the amount of resources allocated to an application or the running speed of the resources can pave the path to new platform-aware implementations. In the first part of the thesis, we introduce a model which takes into account both the topology and the heterogeneity of a platform by introducing two kind of machines. We augment the scheduling problem with constraints whose purpose is to implicitly reduce data movement either during parallel execution or during the communication with the file system. We propose algorithms that can decide the number of resources allocated to an application taking into consideration the extra constraints.In the second part of the thesis, we deal with the uncertainty on part of the input and more specifically, the workload of an application, that is strictly related to the time needed for its completion. Most works in the literature consider this value known in advance. However, this is rarely the case in real-life systems.In our approach, the given workload is a worst case scenario for the execution of an application. We introduce application-specific tests that may decrease the workload of a task.Since the test (e.g. compression) takes some time, and since the amount of reduction (e.g. in size) is unknown before the completion of the test, the decision of running the test for a task or not has to be taken. We propose competitive algorithms for the problem of scheduling such tasks, in order to minimize the energy consumed in a set of speed-adjustable machines. In the third part of the thesis, we focus on a similar setting of uncertain input and we consider a model where the processing times are not known in advance. Here, we augment the input of the problem by introducing predicted values in place of the unknown processing times. We design algorithms that perform optimally when the predictions are accurate while remaining competitive to the best known ones otherwise
Douchet, Fabien. "Optimisation énergétique de data centers par utilisation de liquides pour le refroidissement des baies informatiques." Thesis, Lorient, 2015. http://www.theses.fr/2015LORIS386/document.
Full textData centers are facilities that house a large numbers of computer equipment. More than 99% of the electrical power consumed by the electronic components is converted into heat. To ensure their good working, it is necessary to keep them under their recommended temperatures. This is mainly achieved by the use of air conditioning systems which consume a lot of electrical power. In addition, the power density of computer racks is constantly increasing. So the limits of air as a coolant for electronic equipment cooling are reached.Studies conducted during this thesis concern the improvement of energy efficiency of cooling systems for electronic rack by using liquids as heat transfer fluids. This approach gives higher heat exchange coefficients and larger cooling capacity with more viable aspects for the recovering of heat from data centers.Four cooling solutions are evaluated. Experiments are conducted on several servers and on a computer rack. A consistent instrumentation helps to highlight the efficiency of components cooling and allows us to identify energy efficiency indicators of the studied systems. From the experimental results, two numerical models are developed by a nodal approach and a parameter identification by inverse method is carried out. These models can be duplicated at the scale of a data center room in order to quantify the potential gains of two liquid cooling solutions
Guzzo, Natale. "Facing the real challenges in wireless sensor network-based applications : an adaptative cross-layer self-organization WSN protocol." Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10190.
Full textWireless Sensor Networks (WSN) is one of the protagonists contributing to the evolution and the development of the Internet of Things (IoT). Several use cases can be found today in the different fields of the modern technology including the container shipping industry where containerized cargo accounts for about 60 percent of all world seaborne trade. In this context, TRAXENS developed a battery-powered device named TRAX-BOX designed to be attached to the freight containers in order to track and monitor the shipping goods along the whole supply chain. In this thesis, we present a new energy-efficient self-organizing WSN protocol stack named TRAX-NET designed to allow the TRAX-BOX devices to cooperate to deliver the sensed data to the TRAXENS platform.The results of simulations and field tests show that TRAX-NET well perform in the different scenarios in which it is supposed to operate and better fulfil the requirements of the assumed application in comparison with the existing schemes
Wanza, Weloli Joël. "Modélisation, simulation de différents types d’architectures de noeuds de calcul basés sur l’architecture ARM et optimisés pour le calcul haute-performance." Thesis, Université Côte d'Azur (ComUE), 2019. http://www.theses.fr/2019AZUR4042.
Full textThis work is part of a family of European projects called Mont-Blanc whose objective is to develop the next generation of Exascale systems. It addresses specifically the issue of energy efficiency, at micro-architectural level first by considering the use of 64-bit Armv8-A based compute nodes and an associated relevant SoC topology, and examine also the runtime aspects with notably the study of power management strategies that can be better suited to the constraints of HPC highly parallel processing. A design space exploration methodology capable of supporting the simulation of large manycore computing clusters is developped and lead to propose, design and evaluate multi-SoC and their associated SoC Coherent Interconnect models (SCI). This approach is then used to define a pre-exascale architecture allowing to globally reduce the complexity and cost of chip developments without sacrifying performances. The resulting partitioning scheme introduces interesting perspectives at technology level such as the integration of more compute nodes directly on an interposer based System-in-Package (SiP), possibly based on 3D Through Silicon Vias (TSVs) using High Memory Bandwidth (HBM). Energy efficiency is addressed more directly in second instance by studying current power management policies and proposing two strategies to help reducing power while preserving performances. The first one exploits finer application execution knowledge to adjust the frequency of extensive parallel threads and better balance their execution time. The second strategy reduces core frequencies at synchronisation points of jobs to avoid running the cores at full speed while it is not necessary. Experiment results with these strategies, both in simulation and real hardware, show the possibilities offered par this approach to address the strong requirements of Exascale platforms
Ghribi, Chaima. "Energy efficient resource allocation in cloud computing environments." Thesis, Evry, Institut national des télécommunications, 2014. http://www.theses.fr/2014TELE0035/document.
Full textCloud computing has rapidly emerged as a successful paradigm for providing IT infrastructure, resources and services on a pay-per-use basis over the past few years. As, the wider adoption of Cloud and virtualization technologies has led to the establishment of large scale data centers that consume excessive energy and have significant carbon footprints, energy efficiency is becoming increasingly important for data centers and Cloud. Today data centers energy consumption represents 3 percent of all global electricity production and is estimated to further rise in the future. This thesis presents new models and algorithms for energy efficient resource allocation in Cloud data centers. The first goal of this work is to propose, develop and evaluate optimization algorithms of resource allocation for traditional Infrastructutre as a Service (IaaS) architectures. The approach is Virtual Machine (VM) based and enables on-demand and dynamic resource scheduling while reducing power consumption of the data center. This initial objective is extended to deal with the new trends in Cloud services through a new model and optimization algorithms of energy efficient resource allocation for hybrid IaaS-PaaS Cloud providers. The solution is generic enough to support different type of virtualization technologies, enables both on-demand and advanced resource provisioning to deal with dynamic resource scheduling and fill the gap between IaaS and PaaS services and create a single continuum of services for Cloud users. Consequently, in the thesis, we first present a survey of the state of the art on energy efficient resource allocation in cloud environments. Next, we propose a bin packing based approach for energy efficient resource allocation for classical IaaS. We formulate the problem of energy efficient resource allocation as a bin-packing model and propose an exact energy aware algorithm based on integer linear program (ILP) for initial resource allocation. To deal with dynamic resource consolidation, an exact ILP algorithm for dynamic VM reallocation is also proposed. This algorithm is based on VM migration and aims at constantly optimizing energy efficiency at service departures. A heuristic method based on the best-fit algorithm has also been adapted to the problem. Finally, we present a graph-coloring based approach for energy efficient resource allocation in the hybrid IaaS-PaaS providers context. This approach relies on a new graph coloring based model that supports both VM and container virtualization and provides on-demand as well as advanced resource reservation. We propose and develop an exact Pre-coloring algorithm for initial/static resource allocation while maximizing energy efficiency. A heuristic Pre-coloring algorithm for initial resource allocation is also proposed to scale with problem size. To adapt reservations over time and improve further energy efficiency, we introduce two heuristic Re-coloring algorithms for dynamic resource reallocation. Our solutions are generic, robust and flexible and the experimental evaluation shows that both proposed approaches lead to significant energy savings while meeting the users' requirements
Bonamy, Robin. "Modélisation, exploration et estimation de la consommation pour les architectures hétérogènes reconfigurables dynamiquement." Phd thesis, Université Rennes 1, 2013. http://tel.archives-ouvertes.fr/tel-00931849.
Full textCuadrado-Cordero, Ismael. "Microclouds : an approach for a network-aware energy-efficient decentralised cloud." Thesis, Rennes 1, 2017. http://www.theses.fr/2017REN1S003/document.
Full textThe current datacenter-centralized architecture limits the cloud to the location of the datacenters, generally far from the user. This architecture collides with the latest trend of ubiquity of Cloud computing. Also, current estimated energy usage of data centers and core networks adds up to 3% of the global energy production, while according to latest estimations only 42,3% of the population is connected. In the current work, we focused on two drawbacks of datacenter-centralized Clouds: Energy consumption and poor quality of service. On the one hand, due to its centralized nature, energy consumption in networks is affected by the centralized vision of the Cloud. That is, backbone networks increase their energy consumption in order to connect the clients to the datacenters. On the other hand, distance leads to increased utilization of the broadband Wide Area Network and poor user experience, especially for interactive applications. A distributed approach can provide a better Quality of Experience (QoE) in large urban populations in mobile cloud networks. To do so, the cloud should confine local traffic close to the user, running on the users and network devices. In this work, we propose a novel distributed cloud architecture based on microclouds. Microclouds are dynamically created and allow users to contribute resources from their computers, mobile and network devices to the cloud. This way, they provide a dynamic and scalable system without the need of an extra investment in infrastructure. We also provide a description of a realistic mobile cloud use case, and the adaptation of microclouds on it. Through simulations, we show an overall saving up to 75% of energy consumed in standard centralized clouds with our approach. Also, our results indicate that this architecture is scalable with the number of mobile devices and provide a significantly lower latency than regular datacenter-centralized approaches. Finally, we analyze the use of incentives for Mobile Clouds, and propose a new auction system adapted to the high dynamism and heterogeneity of these systems. We compare our solution to other existing auctions systems in a Mobile Cloud use case, and show the suitability of our solution
Dominici, Michele. "Contributing to energy efficiency through a user-centered smart home." Phd thesis, Université Rennes 1, 2013. http://tel.archives-ouvertes.fr/tel-00869455.
Full textDeddy, Bezeid. "Conception thermique d’une paroi complexe de datacentre pour une optimisation énergétique." Lorient, 2012. http://www.theses.fr/2012LORIS274.
Full textThe reduction of energy consumption for telecommunication buildings is an international challenge for main telecommunication operators and principal actors of internet. Indeed, in these buildings there are electronics equipment with a strong power density and thus a very important thermal contribution. Therefore it is necessary to use large air-conditioning systems in order to maintain ambient conditions (temperature and relative humidity of the air) in fixed ranges. One possible approach for limit installed air conditioning systems is to clip the peaks of internal temperature by using a heat storage in the wall and by adopting a night cool storage directly in the masonry. In this thesis, the study describes a numerical and experimental study in order to define new conceptions of optimized telecommunication buildings. Walls are used in order to increased heat transfer and reduced cooling energy consumption. In the first step, the temperature response of the internal volume 1 m3 were followed and simulated under different test conditions. The thermal inertia is increased by incorporating phase change materials (PCM microencapsulated paraffin) in concrete. From experience and measurements of thermophysical properties, a one dimensional thermal model conduction that represents heat transfer in the walls were developed and validated. From these studies, a specific component, representative of a multilayer wall with PCM, is developed and coupled to TRNSYS Type 56. These TRNSYS developments are then applied to the study of a real "data center" site. After confrontation with experimental data, different configurations of walls have been studied in order to improve thermal inertia. New building architectures are proposed in order to reduce cooling energy consumption
Kaced, Yazid. "Études du refroidissement par free cooling indirect d’un bâtiment exothermique : application au centre de données." Thesis, Lorient, 2018. http://www.theses.fr/2018LORIS499/document.
Full textA data center is a warehouse that contains telecommunication equipment, network infrastructure, servers, and computers. This equipment leads to a very high heat dissipation which must be compensated by the use of cooling systems. Telecommunication standards impose restricted climatic ranges (temperatures and humidity) leading to a very high energy consumption devote to air conditioning. The reduction of this energy consumption constitutes a real challenge which should be raised and solved. Many cooling solutions are proposed as the free cooling solution, which consists in cooling equipment by using external air in propitious climatic conditions. The work carried out during this thesis is based on experiments conducted within a building in real climatic conditions in order to study the cooling of telecom cabinets. During this study, the building configuration was modified, an indirect "free cooling" system was set up and a significant instrumentation was implemented. The objectives are to establish performance factors issued from measurements, to develop and to validate a numerical model in order to predict the thermoaeraulic behavior for this type of solution. Initially, experiments are carried out with a power dissipated inside the building and a cooling provided only by an outside air circulation. Then, significant modifications were made into the building to introduce an internal air circulation in a closed loop in order to evacuate the heat dissipated inside cabinets by a crossing airflow. In order to get a convincing database, measurements were conducted by using one and then several cabinets in different conditions. Modifications are made to operating parameters in order to better understand the installation operation and to define the energy optimization parameters. Numerical models are developed through TRNSYS / TRNFLOW. The confrontation of simulations with measurements shows the implemented approach relevance
Guyot, Dimitri. "Evaluation sur modèle de simulation thermique dynamique calibré des performances d’un contrôleur prédictif basé sur l’utilisation de réseaux de neurones." Thesis, Paris, HESAM, 2020. http://www.theses.fr/2020HESAC022.
Full textThe development of machine learning techniques, particularly neural networks, combined with the development of new information and communication technologies, is shaking up our societies through technological advances in a variety of sectors. The building sector is not spared, so these techniques may represent an interesting opportunity in a context where greenhouse gas emissions must be drastically reduced. The objective of this work is to assess the interest of these techniques in the field of building energy, with the aim of reducing energy consumption and improving thermal comfort. In addition, we ensure that this evaluation is carried out with a global vision, by placing the possible advantages in front of the different needs relating to the development of these technologies. This thesis work is organized in three parts preceded by a detailed introduction intended to give the reader an overview of the various contextual elements, thus allowing the thesis work to be placed in perspective. We then give in the first part the theoretical framework needed to understand the problems encountered during the elaboration and creation of neural networks for building energy applications. Then, a bibliographical study giving the reader a broad overview of the various applications of neural networks in the field of building energy is presented. The second part is devoted to the calibration of the building model that is then used to test and evaluate a predictive controller implementing neural networks. After an explanation of the method used and a detailed presentation of the model, a complete analysis of the calibration results is carried out. We conclude this part with observations and recommendations regarding the standard calibration guidelines recommended by three international organizations. Finally, a practical application using neural networks for the predictive control of indoor temperature is presented in the third part. After a theoretical introduction concerning predictive control, we detail the method employed to train the neural networks used. The results obtained in simulation with a predictive controller are then analyzed and compared with those obtained with two reference controllers for various simulation hypothesis. The predictive controller is thus tested in several scenarios, ranging from an ideal situation to more realistic operating conditions, including two different types of heat emitters, namely radiant ceilings and underfloor heating
Harrane, Ibrahim El Khalil. "Estimation distribuée respectueuse de la consommation d’énergie et de la confidentialité sur les réseaux adaptatifs." Thesis, Université Côte d'Azur (ComUE), 2019. http://www.theses.fr/2019AZUR4041.
Full textDistributed estimation over adaptive networks takes advantage of the interconnections between agents to perform parameter estimation from streaming data. Compared to their centralized counterparts, distributed strategies are resilient to links and agents failures, and are scalable. However, such advantages do not come without a cost. Distributed strategies require reliable communication between neighbouring agents, which is a substantial burden especially for agents with a limited energy budget. In addition to this high communication load, as for any distributed algorithm, there may be some privacy concerns particularly for applications involving sensitive data. The aim of this dissertation is to address these two challenges. To reduce the communication load and consequently the energy consumption, we propose two strategies. The first one involves compression while the second one aims at limiting the communication cost by sparsifying the network. For the first approach, we propose a compressed version of the diffusion LMS where only some random entries of the shared vectors are transmitted. We theoretically analyse the algorithm behaviour in the mean and mean square sense. We also perform numerical simulations that confirm the theoretical model accuracy. As energy consumption is the main focus, we carry out simulations with a realistic scenario where agents turn on and off to save energy. The proposed algorithm outperforms its state of the art counterparts. The second approach takes advantage of the multitask setting to reduce the communication cost. In a multitask setting it is beneficial to only communicate with agents estimating similar quantities. To do so, we consider a network with two types of agents: cluster agents estimating the network structure, and regular agents tasked with estimating their respective objective vectors. We theoretically analyse the algorithm behaviour under two scenarios: one where all agents are properly clustered, and a second one where some agents are asigned to wrong clusters. We perform an extensive numerical analysis to confirm the fitness of the theoretical models and to study the effect of the algorithm parameters on its convergence. To address the privacy concerns, we take inspiration from differentially private Algorithms to propose a privacy aware version of diffusion LMS. As diffusion strategies relies heavily on communication between agents, the data are in constant jeopardy. To avoid such risk and benefit from the information exchange, we propose to use Wishart matrices to corrupt the transmitted data. Doing so, we prevent data reconstruction by adversary neighbours as well as external threats. We theoretically and numerically analyse the algorithm behaviour. We also study the effect of the rank of the Wishart matrices on the convergence speed and privacy preservation
Wambecke, Jérémy. "Visualisation de données temporelles personnelles." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM051/document.
Full textThe production of energy, in particular the production of electricity, is the main responsible for the emission of greenhouse gases at world scale. The residential sector being the most energy consuming, it is essential to act at a personal scale to reduce these emissions. Thanks to the development of ubiquitous computing, it is now easy to collect data about the electricity consumption of electrical appliances of a housing. This possibility has allowed the development of eco-feedback technologies, whose objective is to provide to consumers a feedback about their consumption with the aim to reduce it. In this thesis we propose a personal visualization method for time-dependent data based on a what if interaction, which means that users can apply modifications in their behavior in a virtual way. Especially our method allows to simulate the modification of the usage of electrical appliances of a housing, and then to evaluate visually the impact of the modifications on data. This approach has been implemented in the Activelec system, which we have evaluated with users on real data.We synthesize the essential elements of conception for eco-feedback systems in a state of the art. We also outline the limitations of these technologies, the main one being the difficulty faced by users to find relevant modifications in their behavior to decrease their energy consumption. We then present three contributions. The first contribution is the development of a what if approach applied to eco-feedback as well as its implementation in the Activelec system. The second contribution is the evaluation of our approach with two laboratory studies. In these studies we assess if participants using our method manage to find modifications that save energy and which require a sufficiently low effort to be applied in reality. Finally the third contribution is the in-situ evaluation of the Activelec system. Activelec has been deployed in three private housings and used for a duration of approximately one month. This in-situ experiment allows to evaluate the usage of our approach in a real domestic context. In these three studies, participants managed to find modifications in the usage of appliances that would savea significant amount of energy, while being judged easy to be applied in reality.We also discuss of the application of our what if approach to the domain of personal visualization, beyond electricity consumption data, which is defined as the visual analysis of personal data. We hence present several potential applications to other types of time-dependent personal data, for example related to physical activity or to transportation. This thesis opens new perspectives for using a what if interaction paradigm for personal visualization
Danilo, Robin. "Approches connexionnistes pour la vision par ordinateur embarquée." Thesis, Lorient, 2018. http://www.theses.fr/2018LORIS518/document.
Full textTo design embedded computer vision systems, two axes can be considered. The first focuses on designing new, more powerful, digital devices that can efficiently implement complex algorithms. The second targets the development of new, lightweight computer vision algorithms that can be effectively implemented on digital embedded systems. In this work, we favor the second axis by using connectionist models. In this context, we focus on two models of artificial neural networks: cluster-based networks and convolutional networks. The first model we use, i.e. cluster-based network, was never been used to perform computer vision tasks before. However, it seemed to be a good candidate to design embedded systems, especially through dedicated hardware architectures implementation. The goal was first to find out the kinds of tasks that could be performed using this network model. This model has been designed to implement associative memories which can come close to problems such as content- based image retrieval in computer vision domain. This type of application massively uses approximated nearest neighbor search algorithms which makes it a good candidate to focus on. The second type of network studied in this work, called convolutional network, is very popular to design computer vision systems. Our goal here was to find different ways to simplify their complexity while maintaining high performance. In particular, we proposed a technique that involves re-training quantified networks
Das, Satyajit. "Architecture and Programming Model Support for Reconfigurable Accelerators in Multi-Core Embedded Systems." Thesis, Lorient, 2018. http://www.theses.fr/2018LORIS490/document.
Full textEmerging trends in embedded systems and applications need high throughput and low power consumption. Due to the increasing demand for low power computing and diminishing returns from technology scaling, industry and academia are turning with renewed interest toward energy efficient hardware accelerators. The main drawback of hardware accelerators is that they are not programmable. Therefore, their utilization can be low is they perform one specific function and increasing the number of the accelerators in a system on chip (SoC) causes scalability issues. Programmable accelerators provide flexibility and solve the scalability issues. Coarse-Grained Reconfigurable Array (CGRA) architecture consisting of several processing elements with word level granularity is a promising choice for programmable accelerator. Inspired by the promising characteristics of programmable accelerators, potentials of CGRAs in near threshold computing platforms are studied and an end-to-end CGRA research framework is developed in this thesis. The major contributions of this framework are: CGRA design, implementation, integration in a computing system, and compilation for CGRA. First, the design and implementation of a CGRA named Integrated Programmable Array (IPA) is presented. Next, the problem of mapping applications with control and data flow onto CGRA is formulated. From this formulation, several efficient algorithms are developed using internal resources of a CGRA, with a vision for low power acceleration. The algorithms are integrated into an automated compilation flow. Finally, the IPA accelerator is augmented in PULP - a Parallel Ultra-Low-Power Processing-Platform to explore heterogeneous computing
Hossain, Mohaimenul. "Green Metrics to Improve Sustainable Networking." Electronic Thesis or Diss., Université de Lorraine, 2019. http://www.theses.fr/2019LORR0201.
Full textAchieving energy efficiency has in recent times become a major concern of networking research due to the ever-escalating power consumption and CO2 emissions produced by large data networks. This problem is becoming more and more challenging because of the drastic traffic increase in the last few years and it is expected to be increased even more in the coming years. Using efficient energy-aware strategies that could overturn this situation by reducing the electricity consumption as well as mitigating the environmental impact of data transmission networks. However, CO2 and energy consumption cannot be considered proportionate if the means of electricity production differs. This research work focuses on reducing the environmental impact of data transmission network by implementing energy aware routing, where unused network devices will be put into sleep/shut down and high capacity links will be adapted according to demand requirement. But, alongside with energy, this work has introduced two different metrics namely carbon emission factor and non-renewable energy usage percentage, which are considered as objective functions for designing green network. Here a centralized approach like using Software-Defined Networking (SDN), is used for designing to solve this problem as they allow flexible programmability suitable for this problem. Our proposal proposes a routing technique using genetic algorithm that minimizes the number of network-elements required and at the same time adapt the bandwidth capacity while satisfying an incoming traffic load. Different from existing related works, we focus on optimizing not only energy consumption but also carbon emission and non-renewable energy consumption in SDN in order to close this important gap in the literature and provide solutions compatible with operational backbone networks. Complementing the general aim of improving the environmental impact of data transmission network, this research is also intended to cover important related features such as realistic large demand size, network performance, and Quality of Service (QoS) requirements. At the same time this work focuses on network stability and analyzes the impact of network stability while implementing a green solution. Our work proposes a penalty and filtering mechanism which helps to find an optimal balance between stability and green networking. By using realistic input data, significant values of switched-off links and nodes are reached which demonstrate the effectiveness of our algorithm. The obtained result validated the importance of considering environmental factors rather than considering only energy. Results also show the trade-off between environmental and performance concerns, considering a couple of performance indicators. Moreover, it is shown that the penalty and filtering mechanism is an effective approach to avoid incoherent system and improve the stability of the system. As a whole, this conducted research and contributions reported through this manuscript stand as a valuable solution on the road to sustainable networking
Attoue, Nivine. "Use of Smart Technology for heating energy optimization in buildings : experimental and numerical developments for indoor temperature forecasting." Thesis, Lille 1, 2019. http://www.theses.fr/2019LIL1I021/document.
Full textWith the highly developing concerns about the future of energy resources, the optimization of energy consumption becomes a must in all sectors. A lot of research was dedicated to buildings regarding that they constitute the highest energy consuming sector mainly because of their heating needs. Technologies have been improved and several methods are proposed for energy consumption optimization. Energy saving procedures can be applied through innovative control and management strategies. The objective of this thesis is to introduce the smart concept in the building system to reduce the energy consumption, as well as to improve comfort conditions and users’ satisfaction. The study aims to develop a model that makes it possible to predict thermal behavior of buildings. The thesis proposes a methodology based on the selection of pertinent input parameters, after a relevance analysis of a large set of input parameters, for the development of a simplified artificial neural network (ANN) model, used for indoor temperature forecasting. This model can be easily used in the optimal regulation of buildings’ energy devices. The smart domain needs an automated process to understand the buildings’ dynamics and to describe its characteristics. Such strategies are well described using reduced thermal models. Thus, the thesis presents a preliminary study for the generation of an automated process to determine short term indoor temperature prediction and buildings characteristics based on grey-box modeling. This study is based on a methodology capable of finding the most reliable set of data that describes the best the building’s dynamics. The study shows that the most performant order for reduced-models is governed by the dynamics of the collected data used
Molina, Troconis Laudin Alessandro. "Techniques et métriques non intrusives pour caractériser les réseaux Wi-Fi." Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2018. http://www.theses.fr/2018IMTA0091/document.
Full textNowadays, mobile devices are present worldwide, with over 4.40 Billion devices globally. These devices enable users to access the Internet via wireless networks. Different actors (e.g., home users, enterprises) are installing WiFi networks everywhere, without central coordination, creating chaotic deployments. As a result, WiFi networks are widely deployed all over the world, with high accesspoint (AP) density in urban areas. In this context, end-users and operators are trying to exploit these dense network deployments to obtain ubiquitous Internet connectivity, and possibly other services. However, taking advantage of these deployments requires strategies to gather and provide information about the available networks. In this dissertation, we first study the network discovery process within the context of these deployments. Then, we present the Wireless Measurements Sharing Platform, a collaborative information system, to which mobile stations send simple network measurements that they collected. By gathering and processing several network measurements from different users, the platform provides access to valuable characteristics of the deployment. We evaluate the usefulness of this collaborative platform thanks to two applications: (1) the minimal access point set, to reduce the energy needed to offer WiFi coverage in a given area.(2) The optimization of the scanning parameters,to reduce the time a mobile station needs for the network discovery. Finally, we describe a method to identify whether an AP operates ina saturated channel, by passively monitoring beacon arrival distribution
Ndour, Geneviève. "Approximate computing for high energy-efficiency in IoT applications." Thesis, Rennes 1, 2019. http://www.theses.fr/2019REN1S033/document.
Full textReduced width units are ones of the power reduction methods. However such units have been mostly evaluated separately, i.e. not evaluated in a complete applications. In this thesis, we extend the RISC-V processor with reduced width computation and memory units, in which only a number of most significant bits (MSBs), configurable at runtime is active. The energy reduction vs quality of output trade-offs of applications executed with the extended RISC-V are studied. The results indicate that the energy can be reduced by up to 14% for an error ≤ 0.1%. Moreover we propose a generic energy model that includes both software parameters and hardware architecture ones. It allows software and hardware designers to have an early insight into the effects of optimizations on software and/or units
Gbaguidi, Fréjus A. Roméo. "Approche prédictive de l'efficacité énergétique dans les Clouds Datacenters." Thesis, Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1163/document.
Full textWith the democratization of digital technologies, the construction of a globalized cyberspace insidiously transforms our lifestyle. Connect more than 4 billion people at high speed, requires the invention of new concept of service provision and trafic management that are capable to face the challenges. For that purpose, Cloud Computing have been set up to enable Datacenters to provide part or total IT components needed by companies for timely services delivering with performance that meets the requirements of their clients. Consequently, the proliferation of Datacenters around the world has brought to light the worrying question about the amount of energy needed for their function and the resulting difficulty for the humanity, whose current reserves are not extensible indefinitely. It was therefore necessary to develop techniques that reduce the power consumption of Datacenters by minimizing the energy losses orchestrated on servers where each wasted watt results in a chain effect on a substantial increase in the overall bill of Datacenters. Our work consisted first in making a review of the literature on the subject and then testing the ability of some prediction tools to improve the anticipation of the risks of energy loss caused by the misallocation of virtual equipment on servers. This study focused particularly on the ARMA tools and neural networks which in the literature have produced interesting results in related fields. After this step, it appeared to us that ARMA tools, although having less performance than neural networks in our context, runs faster and are best suited to be implemented in cloud computing environments. Thus, we used the results of this method to improve the decision-making process, notably for the proactive re-allocation of virtual equipment before it leads to under-consumption of resources on physical servers or over-consumption inducing breaches of SLAs. Based on our simulations, this approach enabled us to reduce energy consumption on a firm of 800 servers over a period of one day by more than 5Kwh. This gain could be significant when considering the enormous size of modern data centers and projected over a relatively long period of time. It would be even more interesting to deepen this research in order to generalize the integration of this predictive approach into existing techniques in order to significantly optimize the energy consumption within Datacenters while preserving performance and quality of service which are key requirements in the concept of Cloud Computing
Zabada, Shaker. "Analysis of heating expenditure in social housing : application of economic provisional models." Thesis, Lille 1, 2014. http://www.theses.fr/2014LIL10181/document.
Full textThe research conducted in this doctoral thesis concerns a major socio-economic issue, that of the heating consumption in social housing. It aims at understanding the influence of both building characteristics as well as socio-economic indicators on the heating consumption in this sector and the development of numerical models for the prediction of this consumption. The research is based on data provided by Lille Métropole Habitat, who is in charge of the management of a large social housing stock in Lille Metropolis. The thesis includes four parts. The first part presents a literature review which covers the social housing in Europe, in particular in France, the factors affecting the energy consumption in social housing, and policies proposed for the energy saving in this sector. The second part presents the data used in this work that are provided by LMH. The data concern a large social housing stock in Lille Metropolis (North of France). They include heating expenses as well as the buildings characteristics and some socio-economic indicators on the tenants. The third part presents analysis of the influence of both building characteristics (age, DPE, dwellings’ area, number of floors) and socio-economic parameters (tenants’ age, marital status and income) on the heating consumption. The last part presents the elaboration of prediction models for the heating expenses in the LMH housing stock and the use of these models to analyze the investment policy in the renovation of this stock. Two methods are used: the classical Ordinary Least Squares method (OLS) and the Artificial Neural Networks
Randriatsiferana, Rivo Sitraka A. "Optimisation énergétique des protocoles de communication des réseaux de capteurs sans fil." Thesis, La Réunion, 2014. http://www.theses.fr/2014LARE0019/document.
Full textTo increase the lifetime of wireless sensor networks, a solution is to improve the energy efficiency of the communication's protocol. The grouping of nodes in the wireless sensor network clustering is one of the best methods. This thesis proposes several improvements by changing the settings of the reference protocol LEACH. To improve the energy distribution of "cluster-heads", we propose two centralized clustering protocols LEACH and k-optimized version k-LEACH-VAR. A distributed algorithm, called e-LEACH, is proposed to reduce the periodic exchange of information between the nodes and the base station during the election of "cluster-heads". Moreover, the concept of energy balance is introduced in metric election to avoid overloading nodes. Then we presented a decentralized version of k-LEACH, which in addition to the previous objectives, integrates the overall energy consumption of the network. This protocol, called k-LEACH-C2D, also aims to promote the scalability of the network. To reinforce the autonomy and networks, both routing protocols "multi-hop" probability, denoted CB-RSM and FRSM build elementary paths between the "cluster-heads" and elected the base station. The protocol, CB-RSM, forms a hierarchy of "cluster-heads" during the training phase clusters, with an emphasis on self-scheduling and self-organization between "cluster-heads" to make the networks more scalable. These protocols are based on the basic idea that the nodes have the highest residual energy and lower variance of energy consumption become "cluster-head". We see the central role of consumption of the node in our proposals. This point will be the last part of this thesis. We propose a methodology to characterize experimentally the consumption of a node. The objectives are to better understand the consumption for different sequences of the node status. In the end, we propose a global model of the consumption of the node