Idag är det vanligt att hantera, beskriva och konfigurera sin datainfrastruktur såsom processer, serverar och miljöer i maskinläsbara konfigurationsfiler istället för fysisk hårdvara eller interaktiva konfigureringsverktyg. Automatiserad datainfrastruktur blir mer och mer vanligt för att kunna fokusera mer på utveckling och samtidigt få ett stabilare system. Detta har gjort att antalet verktyg för automatisering av datainfrastruktur skjutit i höjden det senaste årtiondet. Lösningar för automatisering av olika typer av datainfrastrukturer har blivit mer komplexa och innehåller ofta många verktyg som interagerar med varandra. Det här kandidatarbetet jämför, väljer ut och sätter ihop existerande plattformar och verktyg och skapar ett privat moln som infrastruktur för utveckling. Detta för att effektivera livscykeln för en serverbaserad runtime-miljö. En jämförelse av molnplattformarna OpenStack, OpenNebula, CloudStack och Eucalyptus baserad på litteratur, lägger grunden för molnet. Molnplattformen kompletteras därefter med andra verktyg och lösningar för att fullborda livscykelautomatiseringen av runtime-miljöer. En prototyp av lösningen skapades för att analysera praktiska problem. Arbetet visar att en kombination av OpenStack, Docker, containerorkestrering samt konfigureringsverktyg är en lovande lösning. Lösningen skalar efter behov, automatiserar och hanterar verksamhetens konfigurationer för runtime-miljöer.

Designing, building, and implementing a private cloud hosting solution can be challenging. This report aims to unify research in multiple areas within cloud hosting to simplify the process by presenting a comprehensive ground-up approach. The proposed approach includes methods used to decide which models and paradigms to be used, such as abstraction level and infrastructure scale. A step-by-step guide is presented, with all considerations made along the way. The result is a platform accessible from a web browser or through a command-line interface and hosts services such as servers for machine learning and containerized applications in Kubernetes. Further work includes increasing the abstraction layer and enabling hardware enrollment over the network. Moreover, whether this implementation will scale in an intended way remains to be examined. / Att designa, bygga och implementera en privat plattform för molntjänster kan vara utmanande. Den här rapportens mål är att sammanställa forskning inom flera olika områden av cloud-hosting genom ett omfattande och grundligt tillvägagångsätt. Det förslagna tillvägagångsättet inkluderar metoder för att bestämma vilka modeller och paradigmer som ska användas, såsom abstraktionsnivå och skala av infrastruktur. Rapporten presenterar en guide till processen, med alla överväganden som gjordes längs vägen. Resultatet är en plattform som är tillgänglig från en webbläsare eller via en kommandotolk, och agerar värd för tjänster som servrar för maskininlärning och containeriserade applikationer i Kubernetes. Ytterligare arbete inkluderar att abstrahera bort fler aspekter och möjliggöra registrering av ny hårdvara över nätverket. Det återstår att undersöka om denna implementering kommer kunna skala på tänkt sätt.

Cloud computing is increasingly attracting large attention both in academic research and in industrial initiatives. However, despite the popularity, there is a lack of research on the suitability of software tools and parameters for creating and deploying Cloud test beds. Virtualization and how to set up virtual environments can be done through software tools, which are available as open source, but there still needs to be work in terms of which tools to use and how to monitor parameters with the suitability of hardware resources available. This paper discusses the concepts of virtualization, as a practical view point, presenting an in-depth critical analysis of open source cloud implementation tools such as CloudStack, Eucalyptus, Nimbus, OpenStack, OpenNebula, OpenIoT, to name a few. This paper analyzes the various toolkits, parameters of these tools, and their usability for researchers looking to deploy their own Cloud test beds. The paper also extends further in developing an experimental case study of using OpenStack to construct and deploy a test bed using current resources available in the labs at the University of Bradford. This paper contributes to the theme of software setups and open source issues for developing Cloud test bed for deploying and constructing private Cloud test bed.

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / O escalonamento de recursos à um processo chave para a plataforma de ComputaÃÃo em Nuvem, que geralmente utiliza mÃquinas virtuais (MVs) como unidades de escalonamento. O uso de tÃcnicas de virtualizaÃÃo fornece grande flexibilidade com a habilidade de instanciar vÃrias MVs em uma mesma mÃquina fÃsica (MF), modificar a capacidade das MVs e migrÃ-las entre as MFs. As tÃcnicas de consolidaÃÃo e alocaÃÃo dinÃmica de MVs tÃm tratado o impacto da sua utilizaÃÃo como uma medida independente de localizaÃÃo. à geralmente aceito que o desempenho de uma MV serà o mesmo, independentemente da MF em que ela à alocada. Esta à uma suposiÃÃo razoÃvel para um ambiente homogÃneo, onde as MFs sÃo idÃnticas e as MVs estÃo executando o mesmo sistema operacional e aplicativos. No entanto, em um ambiente de ComputaÃÃo em Nuvem, espera-se compartilhar um conjunto composto por recursos heterogÃneos, onde as MFs podem variar em termos de capacidades de seus recursos e afinidades de dados. O objetivo principal deste trabalho à apresentar uma arquitetura que possibilite a padronizaÃÃo da representaÃÃo do poder de processamento das MFs e MVs, em funÃÃo de Unidades de Processamento (UPs), apoiando-se na limitaÃÃo do uso da CPU para prover isolamento de desempenho e manter a capacidade de processamento das MVs independente da MF subjacente. Este trabalho busca suprir a necessidade de uma soluÃÃo que considere a heterogeneidade das MFs presentes na infraestrutura da Nuvem e apresenta polÃticas de escalonamento baseadas na utilizaÃÃo das UPs. A arquitetura proposta, chamada FairCPU, foi implementada para trabalhar com os hipervisores KVM e Xen, e foi incorporada a uma nuvem privada, construÃda com o middleware OpenNebula, onde diversos experimentos foram realizados para avaliar a soluÃÃo proposta. Os resultados comprovam a eficiÃncia da arquitetura FairCPU em utilizar as UPs para reduzir a variabilidade no desempenho das MVs, bem como para prover uma nova maneira de representar e gerenciar o poder de processamento das MVs e MFs da infraestrutura. / Resource scheduling is a key process for cloud computing platform, which generally uses virtual machines (VMs) as scheduling units. The use of virtualization techniques provides great flexibility with the ability to instantiate multiple VMs on one physical machine (PM), migrate them between the PMs and dynamically scale VMâs resources. The techniques of consolidation and dynamic allocation of VMs have addressed the impact of its use as an independent measure of location. It is generally accepted that the performance of a VM will be the same regardless of which PM it is allocated. This assumption is reasonable for a homogeneous environment where the PMs are identical and the VMs are running the same operating system and applications. Nevertheless, in a cloud computing environment, we expect that a set of heterogeneous resources will be shared, where PMs will face changes both in terms of their resource capacities and as also in data affinities. The main objective of this work is to propose an architecture to standardize the representation of the processing power by using processing units (PUs). Adding to that, the limitation of CPU usage is used to provide performance isolation and maintain the VMâs processing power at the same level regardless the underlying PM. The proposed solution considers the PMs heterogeneity present in the cloud infrastructure and provides scheduling policies based on PUs. The proposed architecture is called FairCPU and was implemented to work with KVM and Xen hypervisors. As study case, it was incorporated into a private cloud, built with the middleware OpenNebula, where several experiments were conducted. The results prove the efficiency of FairCPU architecture to use PUs to reduce VMsâ performance variability, as well as to provide a new way to represent and manage the processing power of the infrastructureâs physical and virtual machines.