The aim of this work is to design data center placed into an ISO container used for transportation. Maximization of the space efficiency and minimization of the running costs are primary. This work also contains a measurement of data center electronic parts power consumption, a computational fluid dynamics model, a simulation of external thermal gains and a 3D model with basic technical drawings.

This Master Thesis deals with the design of part of ICT infrastructure migration of two sites of the international company to the external data center. It analyzes the current status of company ICT, describes the virtualization infrastructure and its operational costs. Main part of the Thesis describes a possible solution of migration of virtualization infrastructure to external data center by using project management tools and methods. Finally, the Thesis summarizes project benefits for the company.

<p>Evaporative cooling is a highly energy efficient alternative to conventional vapor compression cooling system. The sensible cooling effect of evaporative cooling systems is well documented in the literature. Direct evaporative cooling however increases the relative humidity of the air as it cools it. This has made it unsuitable for data centers and other applications where humidity control is important. Desiccant-based dehumidifiers (liquid, solid or composites) absorb moisture from the cooled air to control humidity and is regenerated using waste heat from the data center. This work is an experimental and theoretical investigation of the use of desiccant assisted evaporative cooling for data center cooling according to ASHRAE thermal guidelines, TC 9.9. The thickness (depth) of the cooling pad was varied to study its effect on sensible heat loss and latent heat gain. The velocity of air through the pad was measured to determine its effect on sensible cooling. The flow rate of water over the pad was also varied to find the optimal flow for rate for dry bulb depression. The configuration was such that the rotary desiccant wheel (impregnated with silica gel) comes after the direct evaporative cooler. The rotary desiccant wheel was split in a 1:1 ratio for cooling and reactivation at lower temperatures. The dehumidification effectiveness of a fixed bed desiccant dehumidifier was compared with that of a rotary desiccant wheel and a thermoelectric dehumidifier. A novel condensate recovery system using the Peltier effect was proposed to recover moisture from the return air stream, (by cooling the return air stream below its dew point temperature) thereby optimizing the water consumption of evaporative cooling technology and providing suitable air quality for data center cooling. The moisture recovery unit was found to reduce the mass of water lost through evaporation by an average of fifty percent irrespective of the pad depth.</p> <p> </p>

Data centers hosting internet-scale services consume megawatts of power. Mainly for cost reasons but also to appease environmental concerns, data center operators are interested to reduce their use of energy. This thesis investigates if and how hardware virtualization helps to improve the energy efficiency of modern cloud data centers. Our main motivation is to power off unused servers to save energy. The work encompasses three major parts: First, a simulation-driven analysis to quantify the benefits of known reservation times in infrastructure clouds. Virtual machines with similar expiration times are co-located to increase the probability to power down unused physical hosts. Second, we propose and prototyped a system to deliver truly on-demand cloud services. Idle virtual machines are suspended to free resources and as a first step to power off the physical server. Third, a novel block-level data synchronization tool enables fast and efficient state replication. Frequent state synchronization is necessary to prevent data unavailability: powering down a server disables access to the locally attached disks and any data stored on them. The techniques effectively reduce the overall number of required servers either through optimized scheduling or by suspending idle virtual machines. Fewer live servers translate into proportional energy savings, as the unused servers must no longer be powered.

The growing industry of Information and Communication Technology requires higher computing capacity of data centers. The air conditioning in data centers is a key to assure a sustainable computing environment. However, the traditional cooling systems cost large environmental footprints especially on energy consumption and greenhouse gas emissions. As a result, a green innovation of data center cooling solutions is taking place. The telecommunication company Teliasonera is developing a high density data center cooling system - the “Green Room” and has been studying the environmental performance of this system using a Life Cycle approach. As an extension of the previous study, more aspects of the project i.e. the location of the data center, life span, alternative cooling solutions, energy recovery possibilities and uncertainty analysis is explored using Life Cycle Assessment (LCA) methodology. The comparison of locations of the Green Room indicates that the local temperature and electricity production sources are essential factors for the environmental performance of the Green Room. The analysis of the Green Room’s life span reveals that the utilization phase may not always cause the most significant impact during the whole life cycle of the Green Room. If the life span changes, the manufacture phase may predominate the life cycle of the Green Room. The comparative result of alternative cooling technologies addresses that utilizing “natural coolant” (e.g. geo cooling) is a key for sustainable cooling innovation as it could significantly reduce the environmental footprint of the cooling system. Besides, heating a single building (partly) by the waste heat generated from the Green Room could save 30% of cumulative energy input and could reduce more than half of the total environmental impact. Additionally, results uncertainties caused by the choice of different LCIA methods are discussed in the end of the study. / Teliasonera's Green Room concept

The growing industry of Information and Communication Technology requires higher computing capacity of data centers. The air conditioning in data centers is a key to assure a sustainable computing environment. However, the traditional cooling systems cost large environmental footprints especially on energy consumption and greenhouse gas emissions. As a result, a green innovation of data center cooling solutions is taking place. The telecommunication company Teliasonera is developing a high density data center cooling system - the “Green Room” and has been studying the environmental performance of this system using a Life Cycle approach. As an extension of the previous study, more aspects of the project i.e. the location of the data center, life span, alternative cooling solutions, energy recovery possibilities and uncertainty analysis is explored using Life Cycle Assessment (LCA) methodology. The comparison of locations of the Green Room indicates that the local temperature and electricity production sources are essential factors for the environmental performance of the Green Room. The analysis of the Green Room’s life span reveals that the utilization phase may not always cause the most significant impact during the whole life cycle of the Green Room. If the life span changes, the manufacture phase may predominate the life cycle of the Green Room. The comparative result of alternative cooling technologies addresses that utilizing “natural coolant” (e.g. geo cooling) is a key for sustainable cooling innovation as it could significantly reduce the environmental footprint of the cooling system. Besides, heating a single building (partly) by the waste heat generated from the Green Room could save 30% of cumulative energy input and could reduce more than half of the total environmental impact. Additionally, results uncertainties caused by the choice of different LCIA methods are discussed in the end of the study. / Teliasonera's Green Room concept

El presente proyecto de investigación realizado en la empresa Autoespar S.A. contiene un estudio de la infraestructura de servidores y servicio de TI que nos permite buscar la mejor opción tecnológica para el diseño un centro de datos virtual. Esta infraestructura estará acorde a los requerimientos y necesidades que la empresa demanda, con la finalidad de resolver los problemas más críticos, gestionar mejor los recursos, mejorar la disponibilidad de servicios y reducir el tiempo inactividad. El primer capítulo de este proyecto contiene la información de la empresa y la situación problemática que buscamos resolver. Asimismo, definiremos el objetivo principal y los objetivos específicos. El segundo capítulo, encontraremos el marco teórico consistente a nivel teórico y técnico que nos brindara el soporte a todos los puntos para este proyecto. El tercer capítulo nos muestra a detalle el análisis del problema identificado y definimos los requerimientos del proyecto acorde a los objetivos planteados. Finalmente se muestra el registro de interesados y EDT del proyecto. El cuarto capítulo nos muestra el diseño de la solución dividido en dos, el diseño lógico donde se ubicaran los cálculos requeridos para el diseño de la infraestructura virtual. Y el diseño físico, donde se brinda las especificaciones técnicas a considerar para la nueva infraestructura. Finalmente, el quinto capítulo, donde se detalla un plan de pruebas que nos permitirá validar el cumplimento de los objetivos específicos del proyecto a través de un laboratorio de pruebas que demuestra la factibilidad de este proyecto. / This research project carried out in the company Autoespar S.A. contains a study of the infrastructure of servers and IT services that allows us to find the best technological option for the design of a virtual data center. This infrastructure will be in accordance with the requirements and needs that the company demands, in order to solve the most critical problems, better manage resources, improve the availability of services and reduce downtime. The first chapter of this project contains the information of the company and the problematic situation that we seek to solve. Likewise, we will define the main objective and the specific objectives. The second chapter, we will find the consistent theoretical framework at a theoretical and technical level that will provide us with support at all points for this project. The third chapter shows us in detail the analysis of the identified problem and we define the requirements of the project according to the objectives set. Finally the registry of interested parties and WBS of the project is shown. The fourth chapter shows us the design of the solution divided into two, the logical design where the calculations required for the design of the virtual infrastructure are located. And the physical design, where the technical specifications to be considered for the new infrastructure are provided. Finally, the fifth chapter, where a test plan is detailed that will allow us to validate the fulfillment of the specific objectives of the project through a test laboratory that demonstrates the feasibility of this project. / Tesis

El presente trabajo de investigación propone una estrategia para proporcionar una infraestructura escalable en el tiempo y altamente disponible para todos los servicios de la organización. Para el diseño de esta infraestructura virtualizada se realizó un análisis profundo de la problemática actual y de sus necesidades como también de los requerimientos del cliente. En el desarrollo de la solución se tomará en cuenta las recomendaciones y buenas prácticas de los productos de cada fabricante, así como también el uso de la metodología VIM. Primero, expondremos información de la organización que se beneficiara del proyecto, luego se identificara la problemática actual y con los datos obtenidos definiremos los objetivos de estudio. Para finalizar, expondremos cual será la justificación del desarrollo del diseño. Segundo, se desarrollará el marco teórico que incluirá toda la información referente al desarrollo del proyecto y facilitará su comprensión. Tercero, en este capítulo se sustentará el análisis del problema identificado en el primer capítulo con datos reales, luego definiremos la identificación y justificación de los requerimientos del proyecto. Cuarto, se desarrollará las especificaciones técnicas del equipamiento tecnológico que se utilizará en el diseño, se incluirá los cálculos de la cantidad necesaria de los recursos de hardware de memoria RAM, almacenamiento y procesamiento. Quinto, se expondrá los resultados y validaciones que permitirán comprobar el cumplimiento de los objetivos de investigación en función a los indicadores del logro y sus métricas. / This research work proposes a strategy to provide a scalable infrastructure over time and highly available for all the organization's services. For the design of this virtualized infrastructure, a deep analysis of the current problem and its needs as well as the client's requirements was carried out. In developing the solution, the recommendations and good practices of the products of each manufacturer will be taken into account, as well as the use of the VIM methodology. First, we will present information on the organization that will benefit from the project, then the current problem will be identified and with the data obtained we will define the study objectives. Finally, we will expose what will be the justification for the development of the design. Second, the theoretical framework will be developed that will include all the information regarding the development of the project and will facilitate its understanding. Third, this chapter will support the analysis of the problem identified in the first chapter with real data, then we will define the identification and justification of the project requirements. Fourth, the technical specifications of the technological equipment that will be used in the design will be developed, calculations of the necessary amount of RAM memory, storage and processing hardware resources will be included. Fifth, the results and validations will be exposed that will allow to verify the fulfillment of the research objectives based on the achievement indicators and their metrics. / Tesis

Molntjänster är ett snabbt växande område och utgör en viktig funktion i samhället och för verksamheter. Det som driver molntjänster är storskaliga datacenter som samlar tusentals servrar med infrastruktur. Dessa datacenter kräver en hög energianvändning för att drivas vilket utgör ett stort koldioxidutsläpp. Leverantörer upplever ökade krav i hållbarhetsfrågor och arbetar aktivt för att möta dessa krav. Verksamheter har därför en betydande roll i hur leverantörer arbetar med hållbarhet genom att ställa hållbarhetskrav. Det finns däremot utmaningar med att ställa hållbarhetskrav vilket är brist på kunskap, strategier och resurser.  Studien avser att identifiera de ekologiska hållbarhetsaspekter som verksamheter kan kravställa mot molntjänstleverantörer. Syftet med studien var att förmedla kunskap om vilka ekologiska hållbarhetsaspekter som kan appliceras i verksamheters kravspecifikation mot molntjänstleverantörer. Tidigare forskning beskriver hållbarhetsaspekter som molntjänstleverantörer kan applicera för att minska sitt koldioxidutsläpp. Dessa aspekter är: miljöcertifiering, nyckeltal, förnybar energi och energieffektiviserade strategier. Studien antog en kvalitativ forskningsansats. Datainsamlingen utfördes genom semistrukturerade intervjuer. Analysen utfördes med en tematisk analysmetod och det identifierades tre teman vilket är kunskapslucka, prioritering och resursbehov. Studien diskuterar dessa teman i relation till tidigare forskning kring hållbarhetsaspekterna och utmaningar.  Slutsatsen av studien är fem identifierade hållbarhetsaspekter som är: miljöcertifiering, nyckeltal, förnybar energi, geografisk plats och energieffektiviserade strategier. Dessa hållbarhetsaspekter kan verksamheter kravställa mot molntjänstleverantörer och kan reducera utmaningarna med att identifiera hållbarhetskrav. / Cloud services are a rapidly growing field and play a crucial role in society and businesses. What drives cloud services are large-scale data centers that house thousands of servers with infrastructure. These data centers require high energy consumption to operate, resulting in significant carbon emissions. Providers are facing increased demands for sustainability and actively working to meet these requirements. Businesses, therefore, have a significant role in influencing how providers approach sustainability by setting sustainability criteria. However, there are challenges in setting sustainability requirements due to a lack of knowledge, strategies, and resources. This study aims to identify the ecological sustainability aspects that businesses can demand from cloud service providers. The purpose of the study was to provide knowledge about the ecological sustainability aspects that can be included in businesses' specifications when selecting cloud service providers. Previous research describes sustainability aspects that cloud service providers can implement to reduce their carbon emissions. These aspects include environmental certification, key performance indicators, renewable energy, and energy-efficient strategies. The study adopted a qualitative research approach. Data collection was conducted through semi-structured interviews. The analysis was performed using thematic analysis, and three themes were identified: knowledge gap, prioritization, and resource requirements. The study discusses these themes in relation to previous research on sustainability aspects and challenges. The conclusion of the study highlights five identified sustainability aspects: environmental certification, key performance indicators, renewable energy, geographical location, and energy-efficient strategies. These sustainability aspects can be demanded by businesses from cloud service providers and can help address the challenges of identifying sustainability requirements.

The networks in traditional data centers have remained unchanged for decades and have grown large, complex and costly. Many data centers have a general purpose Ethernet network and one or more additional specialized networks for storage or high performance low latency applications. Network convergence promises to lower the cost and complexity of the data center network by virtualizing the different networks onto a single wire. There is little evidence, aside from vendors' claims, that validate network convergence actually achieves these goals. This work defines a framework for creating a series of unbiased tests to validate converged technologies and compare them to traditional configurations. A case study involving two different network converged technologies was developed to validate the defined methodology and framework. The study also shows that these two technologies do indeed perform similarly to non-virtualized network, reduce costs, cabling, power consumption and are easy to operate.