L'expansion des services en ligne, l'avènement du big data, favorisé par l'internet des objets et les terminaux mobiles, a entraîné une croissance exponentielle du nombre de centres de données qui fournissent des divers services de cloud computing. Par conséquent, la topologie du centre de données est considérée comme un facteur d'influence sur la performance du centre de données. En effet, les topologies des centres de données devraient offrir une latence faible, une longueur de chemin moyenne réduite avec une bande passante élevée. Ces exigences augmentent la consommation énergétique dans les centres de données. Dans cette dissertation, différentes solutions ont été proposées pour surmonter ces problèmes. Tout d'abord, nous proposons une nouvelle topologie appelée LCT (Linked Cluster Topology) qui augmente le nombre de nœuds, améliore la connexion réseau et optimise le routage des données pour avoir une faible latence réseau. Une nouvelle topologie appelée VacoNet (Variable connexion Network) a été également présentée. VacoNet offre un nouveau algorithme qui définit le exact nombre de port par commutateur pour connecter le nombre de serveurs requis tout en réduisant l'énergie consommée et les matériaux inutilisés (câbles, commutateurs). En outre, nous _étudions une nouvelle technique pour optimiser la consumation d'énergie aux centres de données. Cette technique peut périodiquement estimer la matrice de trafic et gérer l'_état des ports de serveurs tout en maintenant le centre de données entièrement connecté. La technique proposée prend en considération le trafic réseau dans la décision de gestion des ports. / Data centers (DC) are being built around the world to provide various cloud computing services. One of the fundamental challenges of existing DC is to design a network that interconnects massive number of nodes (servers)1 while reducing DC' cost and energy consumption. Several solutions have been proposed (e.g. FatTree, DCell and BCube), but they either scale too fast (i.e., double exponentially) or too slow. Effcient DC topologies should incorporate high scalability, low latency, low Average Path Length (APL), high Aggregated Bottleneck Throughput (ABT) and low cost and energy consumption. Therefore, in this dissertation, different solutions have been proposed to overcome these problems. First, we propose a novel DC topology called LCT (Linked Cluster Topology) as a new solution for building scalable and cost effective DC networking infrastructures. The proposed topology reduces the number of redundant connections between clusters of nodes, while increasing the numbers of nodes without affecting the network bisection bandwidth. Furthermore, in order to reduce the DCs cost and energy consumption, we propose first a new static energy saving topology called VacoNet (Variable Connection Network) that connects the needed number of servers while reducing the unused materials (cables, switches). Also, we propose a new approach that exploits the correlation in time of internode communication and some topological features to maximize energy saving without too much impacting the average path length.

Atualmente, o processamento de dados é uma operação vital para os negócios de qualquer companhia, indústria ou prestador de serviços. Entretanto, os servidores em operação, dissipam energia térmica, aquecendo o ambiente, podendo ocasionar problemas operacionais, como funcionamento inadequado, paradas operacionais e até incêndios. Por isso, é necessário um sistema de condicionamento de ar para controlar de temperatura e umidade ambiental, entretanto o consumo de energia elétrica deste sistema aumenta o custo operacional da instalação, sendo necessário minimizá-lo. Baseado neste fato, esta tese avalia a aplicação de 2 tipos de sistemas de resfriamento solar, absorção térmica solar e fotovoltaico, para reduzir o consumo do tradicional sistema elétrico com unidades de resfriamento elétricas operadas por compressão de vapor, apresentando economias, disponibilidade diária e vantagens em uma sala de missão crítica localiza genericamente, em São Paulo, Abu Dhabi, Los Angeles e Nova York. Além disto, analisa a contribuição de sistemas de resfriamento natural e seus efeitos sobre carga e consumo. As densidades de carga propostas foram de 0,5 kW?m-², 1,0 kW?m-², 2,0 kW?m-², 4,0 kW?m-² e 8,0 kW/m. Os índices locais de irradiação e temperaturas anuais baseiam-se nos dados apresentados no banco de dados da ASHRAE para as localidades. Os resultados, válidos para um ano típico e são comparados com (1) Sala de missão crítica acionada por um sistema convencional, (2) combinação de métodos de resfriamento solar, (3) combinação de técnicas solares e resfriamento natural. Por meio de estimativas anuais, calculando as contribuições dos métodos de resfriamento solar e natural, concluiu- maioria dos casos o método de absorção solar apresenta maiores vantagens de economia de energia e retorno de investimento que o método fotovoltaico, especialmente quando associado ao resfriamento natural indireto. Ao final, conclui-se que a utilização de resfriamento solar, pode atingir a 50% de economia na conta de energia e um TIR de 25%. Quando associado a sistemas de resfriamento natural, a economia pode atingir valores de 68% de economia e um TIR de 134%. Para tal deverá ser observado a carga aplicada, o valor de investimento e a curva característica da unidade de resfriamento que irá complementar a carga. / Nowadays, data processing is a fundamental operation for modern business such as banks, technology companies, and factories, among others. However, computers dissipate non-negligible amounts of heat and as the operating temperature increases, these machines cannot operate properly or, under extreme conditions, can come to a stop by overheating. Consequently, it is necessary air conditioning systems to keep the proper operating temperature as well as the room temperature itself. On the other hand, a data center air conditioning system drains a large amount of electrical energy in the installation. Based on this, this thesis evaluates two types of solar cooling system, an absorption thermal system, and a photovoltaic system, to support the traditional electric chiller system showing the energy saving, advantage over conventional cooling and day availability for this system. In addition, it analyzed free cooling system contribution and his effects on consumption. A case study is analyzed in a generic data center located in the city of São Paulo, Abu Dhabi, Los Angeles, and New York. At first, the electrical power density simulated by the computers is 0.5 kW/m², 1.0 kW/m² 2.0 kW/m², 4.0 kW/m² and 8 kW/m², that would occur in half loads or in high loads computer installations. Local solar irradiation and temperature indexes are based on the ASHRAE database. The results are valid for a typical year and are compared to (1) a conventional data center; (2) the event combination of solar cooling; (3) the event combination of solar cooling and free cooling. In conclusion, the energy savings, payback and internal rate of return for 10 years are presented. In the main cases, the solar absorption takes advantage than photovoltaic cooling, However, It must be observed the load replicated and the influence on the work curve of the chiller and the beginning investment. In conclusion, the solar cooling, especially solar absorption can achieve great advantage to installation, resulting in 50% on savings in the electric bill a return internal rate for 10 years of 25%. When it is associated with free cooling this savings can achieve 68% and return rate of 134%.

Atualmente, o processamento de dados é uma operação vital para os negócios de qualquer companhia, indústria ou prestador de serviços. Entretanto, os servidores em operação, dissipam energia térmica, aquecendo o ambiente, podendo ocasionar problemas operacionais, como funcionamento inadequado, paradas operacionais e até incêndios. Por isso, é necessário um sistema de condicionamento de ar para controlar de temperatura e umidade ambiental, entretanto o consumo de energia elétrica deste sistema aumenta o custo operacional da instalação, sendo necessário minimizá-lo. Baseado neste fato, esta tese avalia a aplicação de 2 tipos de sistemas de resfriamento solar, absorção térmica solar e fotovoltaico, para reduzir o consumo do tradicional sistema elétrico com unidades de resfriamento elétricas operadas por compressão de vapor, apresentando economias, disponibilidade diária e vantagens em uma sala de missão crítica localiza genericamente, em São Paulo, Abu Dhabi, Los Angeles e Nova York. Além disto, analisa a contribuição de sistemas de resfriamento natural e seus efeitos sobre carga e consumo. As densidades de carga propostas foram de 0,5 kW?m-², 1,0 kW?m-², 2,0 kW?m-², 4,0 kW?m-² e 8,0 kW/m. Os índices locais de irradiação e temperaturas anuais baseiam-se nos dados apresentados no banco de dados da ASHRAE para as localidades. Os resultados, válidos para um ano típico e são comparados com (1) Sala de missão crítica acionada por um sistema convencional, (2) combinação de métodos de resfriamento solar, (3) combinação de técnicas solares e resfriamento natural. Por meio de estimativas anuais, calculando as contribuições dos métodos de resfriamento solar e natural, concluiu- maioria dos casos o método de absorção solar apresenta maiores vantagens de economia de energia e retorno de investimento que o método fotovoltaico, especialmente quando associado ao resfriamento natural indireto. Ao final, conclui-se que a utilização de resfriamento solar, pode atingir a 50% de economia na conta de energia e um TIR de 25%. Quando associado a sistemas de resfriamento natural, a economia pode atingir valores de 68% de economia e um TIR de 134%. Para tal deverá ser observado a carga aplicada, o valor de investimento e a curva característica da unidade de resfriamento que irá complementar a carga. / Nowadays, data processing is a fundamental operation for modern business such as banks, technology companies, and factories, among others. However, computers dissipate non-negligible amounts of heat and as the operating temperature increases, these machines cannot operate properly or, under extreme conditions, can come to a stop by overheating. Consequently, it is necessary air conditioning systems to keep the proper operating temperature as well as the room temperature itself. On the other hand, a data center air conditioning system drains a large amount of electrical energy in the installation. Based on this, this thesis evaluates two types of solar cooling system, an absorption thermal system, and a photovoltaic system, to support the traditional electric chiller system showing the energy saving, advantage over conventional cooling and day availability for this system. In addition, it analyzed free cooling system contribution and his effects on consumption. A case study is analyzed in a generic data center located in the city of São Paulo, Abu Dhabi, Los Angeles, and New York. At first, the electrical power density simulated by the computers is 0.5 kW/m², 1.0 kW/m² 2.0 kW/m², 4.0 kW/m² and 8 kW/m², that would occur in half loads or in high loads computer installations. Local solar irradiation and temperature indexes are based on the ASHRAE database. The results are valid for a typical year and are compared to (1) a conventional data center; (2) the event combination of solar cooling; (3) the event combination of solar cooling and free cooling. In conclusion, the energy savings, payback and internal rate of return for 10 years are presented. In the main cases, the solar absorption takes advantage than photovoltaic cooling, However, It must be observed the load replicated and the influence on the work curve of the chiller and the beginning investment. In conclusion, the solar cooling, especially solar absorption can achieve great advantage to installation, resulting in 50% on savings in the electric bill a return internal rate for 10 years of 25%. When it is associated with free cooling this savings can achieve 68% and return rate of 134%.

Submitted by Erika Demachki (erikademachki@gmail.com) on 2015-02-05T19:59:39Z No. of bitstreams: 2 Dissertação - Pedro Henrique Pires de Castro - 2014.pdf: 1908182 bytes, checksum: edac87bddd8346a2bcce5d9b5f00301d (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Erika Demachki (erikademachki@gmail.com) on 2015-02-05T20:00:48Z (GMT) No. of bitstreams: 2 Dissertação - Pedro Henrique Pires de Castro - 2014.pdf: 1908182 bytes, checksum: edac87bddd8346a2bcce5d9b5f00301d (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-02-05T20:00:48Z (GMT). No. of bitstreams: 2 Dissertação - Pedro Henrique Pires de Castro - 2014.pdf: 1908182 bytes, checksum: edac87bddd8346a2bcce5d9b5f00301d (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2014-08-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Cloud computing is being consolidated as a new distributed systems paradigm, offering computing resources in a virtualized way and with unprecedented levels of flexibility, reliability, and scalability. Unfortunately, the benefits of cloud computing come at a high cost with regard to energy, mainly because of one of its core enablers, the data center. There are a number of proposals that seek to enhance energy efficiency in data centers. However, most of them focus only on the energy consumed by CPU and ignore the remaining hardware, e.g., RAM. In this work, we show the considerable impact that RAM can have on total energy consumption, particularly in servers with large amounts of this memory. We also propose three new approaches for dynamic consolidation of virtual machines (VMs) that take into account both CPU and RAM usage. We have implemented and evaluated our proposals in the CloudSim simulator using real-world traces and compared the results with state-of-the-art solutions. By adopting a wider view of the system, our proposals are able to reduce not only energy consumption but also the number of SLA violations, i.e., they provide a better service at a lower cost. / A computação em nuvem tem levado os sistemas distribuídos a um novo patamar, oferecendo recursos computacionais de forma virtualizada, flexível, robusta e escalar. Essas vantagens, no entanto, surgem juntamente com um alto consumo de energia nos centros de dados, ambientes que podem ter até centenas de milhares de servidores. Existem muitas propostas para alcançar eficiência energética em centros de dados para computação em nuvem. Entretanto, muitas propostas consideram apenas o consumo proveniente do uso de CPU e ignoram os demais componentes de hardware, e.g., RAM. Neste trabalho, mostramos o impacto considerável que RAM pode ter sobre o consumo total de energia, principalmente em servidores com grandes quantidades dessa memória. Também propomos três novas abordagens para consolidação dinâmica de máquinas virtuais, levando em conta tanto o consumo de CPU quanto de RAM. Nossas propostas foram implementadas e avaliadas no simulador CloudSim utilizando cargas de trabalho do mundo real. Os resultados foram comparados com soluções do estado-da-arte. Pela adoção de uma visão mais ampla do sistema, nossas propostas não apenas são capazes de reduzir o consumo de energia como também reduzem violações de SLA, i.e., proveem um serviço melhor a um custo mais baixo.

Lättnadsregeln i 11 kap. 15 § LSE ger möjlighet till avdrag för den el som förbrukats i datorhallar. Datorhallsdefintionen avgränsar skattelättnaden till att enbart gälla vissa datacenter enligt 1 kap. 14 § LSE. Skattelättnaderna innebär ett driftstöd, vilket regleras inom statsstödsreglerna. Generellt är statsstöd förbjudet, men Sverige har använt sig av gruppundantaget GBER för att kunna ge stöd till vissa datorhallar med motivationen att det faller in under GBER:s avsnitt för miljöskydd. Dagens utformning av stödmottagar-formuleringen enligt LSE kopplat till skattelättnadsregeln i 11 kap. 15 § LSE uppfyller inte syftet enligt förarbetena rörande skattelättnaderna för datorhallar. Enligt förarbetena framgår det tydligt att syftet är att även colocation-datorhallar är minst lika konkurrensutsatta och elintensiva som andra datorhallar, och därför bör anses lika berättigade till statsstöd som företagsspecifika datorhallar. I 1 kap. 11b § p. 4 framgår dock att stödmottagaren är elförbrukaren, vilket i colocation-sammanhang med största sannolikhet bör ses som colocation-företagens kunder, inte colocation-företagen själva. Detta skiljer sig gentemot de företagsspecifika datorhallarna, vilka förbrukar elen själva i de egna datorhallarna. SKV:s ställningstagande, vilket tolkar begreppet ”förbrukare av el” som ”den som äger utrustningen som elen förbrukas i”, får anses innebära en korrekt tolkning av lagen, så som den nu är utformad. Detta medför att det krävs en lagändring av vem som är stödmottagare i datorhallssammanhang för att syftena med förarbetet ska kunna uppnås. Omformuleringen kan lämpligen ske genom att begreppet stödmottagare, byts från ”förbrukaren av el” till något som tar sikte på själva datorhallen. I nuläget utgår statsstöd till sådana bolag, oavsett verksamhet, som hyr plats för sina servrar hos colocation-datorhallar som uppnår de tekniska kraven på bl.a. installerad effekt för att räknas som datorhall vars elförbrukning ger rätt till stöd. Det skulle möjligen kunna anses att ett så brett spann av stödmottagare inte stämmer överens med vad Sverige angett till kommissionen vid underrättelserna om statsstöd enligt GBER. Eftersom datorhallarna ger många fördelar för Sverige i form av exempelvis inkomster till statskassan och arbetstillfällen, så är det viktigt att Sverige upprätthåller konkurrenskraftiga förutsättningar för att attrahera fler datorhallsföretag. En del av att upprätthålla konkurrenskraften är att ha konkurrenskraftiga regler, däribland konkurrenskraftig prissättning på elen. Jag finner tre lösningar på stödmottagar-problematiken för colocation-företag. Antingen så får man ansöka om stöd istället för att använda sig av gruppundantaget, för att på så vis känna sig säkrare på att stödet blir korrekt utformat, eller separera colocation-regeln från LSE för att komma bort från stödmottagar-problematiken. Det tredje alternativet, vilket jag anser lättast, är att formulera ett tillägg i LSE, gällande vem som anses vara stödmottagare i datorhallssammanhang, exempelvis kopplat till vem som anses vara ägaren till infrastrukturen i datorhallarna. Finlands motsvarighet till vår lättnadsregel slipper uppdelningsproblematiken mellan företagsspecifika datorhallar och colocation-datorhallar, då de ser datorhallsföretagen som skattskyldiga, och då tillämpar en lägre skattesats för elen som datorhallarna förbrukar. LSE skiljer sig dock på flera sätt från den finska lagen. Detta gör att även om inspiration kan inhämtas kring hur fokus bör flyttas för stödmottagar-rollen i datorhallssammanhang till att inbegripa själva datorhallen snarare än elförbrukaren, så är det inte möjligt att kopiera den finska utformningen rakt av heller.

The complex turbulent flow regimes encountered in many thermal-fluid engineering applications have proven resistant to the effective application of systematic design because of the computational expense of model evaluation and the inherent variability of turbulent systems. In this thesis the integration of the Proper Orthogonal Decomposition (POD) for reduced order modeling of turbulent convection with the application of robust design principles is proposed as a practical design approach. The POD has been used successfully to create low dimensional steady state flow models within a prescribed range of parameters. The underlying foundation of robust design is to determine superior solutions to design problems by minimizing the effects of variation on system performance, without eliminating their causes. The integration of these constructs utilizing the compromise Decision Support Problem (DSP) results in an efficient, effective robust design approach for complex turbulent convective systems. The efficacy of the approach is illustrated through application to the configuration of data center server cabinets. Data centers are computing infrastructures that house large quantities of data processing equipment. The data processing equipment is stored in 2 m high enclosures known as cabinets. The demand for increased computational performance has led to very high power density cabinet design, with a single cabinet dissipating up to 20 kW. The computer servers are cooled by turbulent convection and have unsteady heat generation and cooling air flows, yielding substantial inherent variability, yet require some of the most stringent operational requirements of any engineering system. Through variation of the power load distribution and flow parameters, such as the rate of cooling air supplied, thermally efficient configurations that are insensitive to variations in operating conditions are determined. This robust design approach is applied to three common data center server cabinet designs, in increasing levels of modeling detail and complexity. Results of the application of this approach to the example problems studied show that the resulting thermally efficient configurations are capable of dissipating up to a 50% greater heat load and 15% decrease in the temperature variability using the same cooling infrastructure. These results are validated rigorously, including comparison of detailed CFD simulations with experimentally gathered temperature data of a mock server cabinet. Finally, with the approach validated, augmentations to the approach are considered for multi-scale design, extending approaches domain of applicability.

Time series data arise in many applications, from motion capture, environmental monitoring, temperatures in data centers, to physiological signals in health care. In the thesis, I will focus on the theme of learning and mining large collections of co-evolving sequences, with the goal of developing fast algorithms for finding patterns, summarization, and anomalies. In particular, this thesis will answer the following recurring challenges for time series: 1. Forecasting and imputation: How to do forecasting and to recover missing values in time series data? 2. Pattern discovery and summarization: How to identify the patterns in the time sequences that would facilitate further mining tasks such as compression, segmentation and anomaly detection? 3. Similarity and feature extraction: How to extract compact and meaningful features from multiple co-evolving sequences that will enable better clustering and similarity queries of time series? 4. Scale up: How to handle large data sets on modern computing hardware? We develop models to mine time series with missing values, to extract compact representation from time sequences, to segment the sequences, and to do forecasting. For large scale data, we propose algorithms for learning time series models, in particular, including Linear Dynamical Systems (LDS) and Hidden Markov Models (HMM). We also develop a distributed algorithm for finding patterns in large web-click streams. Our thesis will present special models and algorithms that incorporate domain knowledge. For motion capture, we will describe the natural motion stitching and occlusion filling for human motion. In particular, we provide a metric for evaluating the naturalness of motion stitching, based which we choose the best stitching. Thanks to domain knowledge (body structure and bone lengths), our algorithm is capable of recovering occlusions in mocap sequences, better in accuracy and longer in missing period. We also develop an algorithm for forecasting thermal conditions in a warehouse-sized data center. The forecast will help us control and manage the data center in a energy-efficient way, which can save a significant percentage of electric power consumption in data centers.

Recent years have witnessed a surge in demand for infrastructure and services to cover high demands on processing big chunks of data and applications resulting in a mega Cloud Datacenter. A datacenter is of high complexity with increasing difficulties to identify, allocate efficiently and fast an appropriate host for the requested virtual machine (VM). Establishing a good awareness of all datacenter’s resources enables the allocation “placement” policies to make the best decision in reducing the time that is needed to allocate and create the VM(s) at the appropriate host(s). However, current algorithms and policies of placement “allocation” do not focus efficiently on awareness of the resources of the datacenter, and moreover, they are based on conventional static techniques. Which are adversely impacting on the allocation progress of the policies. This thesis proposes a new Agent-based allocation/placement policy that employs some of the Multi-Agent system features to get a good awareness of Cloud Datacenter resources and also provide an efficient allocation decision for the requested VMs. Specifically, (a) The Multi-Agent concept is used as a part of the placement policy (b) A Contract Net Protocol is devised to establish good awareness and (c) A verification process is developed to fully dimensional VM specifications during allocation. These new results show a reduction in response time of VM allocation and the usage improvement of occupied resources. The proposed Agent-based policy was implemented using the CloudSim toolkit and consequently was compared, based on a series of typical numerical experiments, with the toolkit’s default policy. The comparative study was carried out in terms of the time duration of VM allocation and other aspects such as the number of available VM types and the amount of occupied resources. Moreover, a two-stage comparative study was introduced through this thesis. Firstly, the proposed policy is compared with four state of the art algorithms, namely the Random algorithm and three one-dimensional Bin-Packing algorithms. Secondly, the three Bin-Packing algorithms were enhanced to have a two-dimensional verification structure and were compared against the proposed new algorithm of the Agent-based policy. Following a rigorous comparative study, it was shown that, through the typical numerical experiments of all stages, the proposed new Agent-based policy had superior performance in terms of the allocation times. Finally, avenues arising from this thesis are included.

L’arrivée de la prochaine technologie 5G va permettre la connectivité des billions de terminaux mobiles et donc une énorme augmentation du trafic de données. A cet égard, les fournisseurs des services Cloud doivent posséder les infrastructures physiques capables de supporter cette explosion de trafic. Malheureusement, les architectures filaires conventionnelles des centres de données deviennent staturées et la congestion des équipements d’interconnexion est souvent atteinte. Dans cette thèse, nous explorons une approche récente qui consiste à augmenter le réseau filaire du centre de données avec l’infrastructure sans fil. En effet, nous exploitons une nouvelle technologie émergente, la technologie 60 GHz, qui assure un débit de l’ordre de 7 Gbits/s afin d’améliorer la QoS. Nous concevons une architecture hybride (filaire/sans fil) du réseau de centre de données basée sur : i) le modèle "Cisco’s Massively Scalable Data Center" (MSDC), et ii) le standard IEEE 802.11ad. Dans une telle architecture, les serveurs sont regroupés dans des racks, et sont interconnectés à travers un switch Ethernet, appelé top-of-rack (ToR) switch. Chaque ToR switch possède plusieurs antennes utilisées en parallèle sur différents canaux sans fil. L’objectif final consiste à minimiser la congestion du réseau filaire, en acheminant le maximum du trafic sur les canaux sans fil. Pour ce faire, cette thèse se focalise sur l’optimisation du routage et de l’allocation des canaux sans fil pour les communications inter-rack, au sein d’un centre de données hybride (HDCN). Ce problème étant NP-difficile, nous allons procéder en trois étapes. En premier lieu, on considère le cas des communications à un saut, où les racks sont placés dans le même rayon de transmission. Nous proposons un nouvel algorithme d’allocation des canaux sans fil dans les HDCN, qui permet d’acheminer le maximum des communications en sans-fil, tout en améliorant les performances réseau en termes de débit et délai. En second lieu, nous nous adressons au cas des communications à plusieurs sauts, où les racks ne sont pas dans le même rayon de transmission. Nous allons proposer une nouvelle approche optimale traitant conjointement le problème du routage et de l’allocation de canaux sans fils dans le HDCN, pour chaque communication, dans un mode online. En troisième étape, nous proposons un nouvel algorithme qui calcule conjointement le routage et l’allocation des canaux pour un ensemble des communications arrivant en mode batch (i.e., par lot). En utilisant le simulateur réseau QualNet, considérant toute la pile TCP/IP, les résultats obtenus montrent que nos propositions améliorent les performances comparées aux méthodes de l’état de l’art / The high proliferation of smart devices and online services allows billions of users to connect with network while deploying a vast range of applications. Particularly, with the advent of the future 5G technology, it is expected that a tremendous mobile and data traffic will be crossing Internet network. In this regard, Cloud service providers are urged to rethink their data center architectures in order to cope with this unprecedented traffic explosion. Unfortunately, the conventional wired infrastructures struggle to resist to such a traffic growth and become prone to serious congestion problems. Therefore, new innovative techniques are required. In this thesis, we investigate a recent promising approach that augments the wired Data Center Network (DCN) with wireless communications. Indeed, motivated by the feasibility of the new emerging 60 GHz technology, offering an impressive data rate (≈ 7 Gbps), we envision, a Hybrid (wireless/wired) DCN (HDCN) architecture. Our HDCN is based on i) Cisco’s Massively Scalable Data Center (MSDC) model and ii) IEEE 802.11ad standard. Servers in the HDCN are regrouped into racks, where each rack is equipped with a: i) Ethernet top-of-rack (ToR) switch and ii) set of wireless antennas. Our research aims to optimize the routing and the allocation of wireless resources for inter-rack communications in HDCN while enhancing network performance and minimizing congestion. The problem of routing and resource allocation in HDCN is NP-hard. To deal with this difficulty, we will tackle the problem into three stages. In the first stage, we consider only one-hop inter-rack communications in HDCN, where all communicating racks are in the same transmission range. We will propound a new wireless channel allocation approach in HDCN to hardness both wireless and wired interfaces for incoming flows while enhancing network throughput. In the second stage, we deal with the multi-hop communications in HDCN where communicating racks can not communicate in one single-hop wireless path. We propose a new approach to jointly route and allocate channels for each single communication flow, in an online way. Finally, in the third stage, we address the batched arrival of inter-rack communications to the HDCN so as to further optimize the usage of wireless and wired resources. For that end, we propose: i) a heuristic-based and ii) an approximate, solutions, to solve the joint batch routing and channel assignment. Based on extensive simulations conducted in QualNet simulator while considering the full protocol stack, the obtained results for both real workload and uniform traces, show that our proposals outperform the prominent related strategies

Actualmente, los sectores empresariales optimizan sus costos sin perder eficiencia productiva en Data Center Móviles tercer izando el servicio y adquiriendo esta tecnología ya que la infraestructura montada es de utilidad para todo ámbito empresarial. El presente proyecto trata sobre el servicio de implementación de un nuevo Data Center Contingente tipo Container, en adelante CPD, como consecuencia de las necesidades de mejora centro de los procesos de negocio. El diseño y dimensionamiento de los diferentes componentes cumplen los estándares requeridos en el mercado. Los Outdoor Enclosure Electric Shelter Prefabricados en sí y todos los equipamientos eléctricos utilizados en sus sistemas para la protección, control y supervisión, están construidos de acuerdo a las Normas vigentes de ANSI, NEMA, ASTM, IEEE, ISA, OSHA, los cuales además cuentan con Aprobaciones y Certificación de Calidad de Laboratorios como UL, CSA, SEC o Laboratorios de Control de Producción y de Certificación de Calidad equivalentes. Actualmente las empresas vienen usando enlaces propios y dedicados para manejar su información y conexión con su data center principal y de contingencia. En muchos casos desaprovechando la comunicación entre ellos por lo que se requiere la necesidad de ser más efectivos en el uso de sus recursos de comunicación, por lo cual se debe considerar los siguientes aspectos: • Modelamiento de Tráfico. • Seguridad. • Conmutación de enlaces. Currently, the business sectors optimize their costs without losing production efficiency in Mobile Data Center outsourcing service and acquiring this technology because the infrastructure is mounted useful for all business field. This project deals with the service of implementing a new quota type Data Center Container, hereinafter CPD, following needs improvement center business processes. The design and dimensioning of the different components meet the required standards in the market. The Outdoor Enclosure Electric Shelter Prefabricated itself and all electrical equipment used in systems for the protection, control and monitoring are built according to current standards of ANSI, NEMA, ASTM, IEEE, ISA, OSHA, which also have with Approvals and Quality Certification Laboratories as UL, CSA, SEC or Control Laboratories Production and Quality Certification equivalents. Currently companies are using own links and dedicated to manage their information and connection with your main data center and contingency. In many cases missing the communication between them so the need to be more effective in their use of communication resources is required, so you should consider the following: • Traffic Modeling. • Security. • Switching links.