Cloud computing Infrastructure-as-a-Service (IaaS) has gained momentum in the cloud computing research field due to its ability to provide efficient infrastructures. Cloud Service Providers (CSPs) are striving to offer Quality of Service (QoS)-guaranteed IaaS services while also improving their resource utilization and maximizing profit. In addition, CSPs are challenged by the need to manipulate diverse and heterogeneous resources, realizing multiple objectives for both customers and CSPs, and handling scalability issues. These challenges are the motivations behind this work which aims at developing a multi-layered framework for constructing and managing efficient IaaS. The fundamental layer in this framework, the Virtual Infrastructure (VI) composition layer, is dedicated to composing and delivering VIs as an IaaS service. This framework relies on a preparatory step that is defined when all the available resources in the managed space are collected in a large repository, the Virtual Resource Pool (VRP). The VRP creation process unifies the representation of all the diverse and heterogeneous resources available. Subsequently, the proposed framework performs various resource allocation approaches as working solutions through the VI composition layer. These approaches adopt efficient techniques and methodologies in performing their operations. The working solutions are initiated by designing a composition approach that relies on an ontology-based model representation. The composition approach exploits semantic similarity, closeness centrality, and random walk techniques for efficient resource allocation. As a result, it provides an efficient solution in a reasonable computational time with no guarantee for the optimality of the obtained solutions. To achieve an optimal solution, the composition approach uses a mathematical modeling formulation. In this solution, the concepts of the composition approach have been integrated into a multi-objective Mixed Integer Linear Programming (MILP) model that has been solved optimally. Despite the optimality of the resulting solution, the MILP-based model restricts IaaS resource allocation to a computational running-time challenge, and the issue of limited-size datacenters. To circumvent these issues, a cost-efficient model is proposed. The new model introduces a Column Generation (CG) formulation for the IaaS resource allocation problem in large datacenters acquainted with QoS requirements. Furthermore, this formulation is realistic, adopts large-scale optimization tools that are adequate for large datacenters, and ensures optimal solutions in a reasonable time. However, growing costs in large datacenters in accordance with the growth of recent large-scale application demands, makes large datacenters economically inefficient. Thus, we advocate a distributed framework for IaaS provisioning that guarantees affordable, scalable, and QoS-assured infrastructure for hosting large-scale applications in geo-distributed datacenters. The framework incorporates two decentralized resource allocation approaches, hierarchical and distributed, that use efficient economic models. These approaches are quite promising solutions for the scalability and computational complexity issues of existing centralized approaches. Finally, the cost-efficient model has been extended to fit the distributed infrastructure by considering additional constraints that impact CSP revenue. Simulation results showcase the effectiveness of the presented work along with the potential benefits of the proposed solutions in terms of satisfying the customers’ requirements, while achieving a better resource utilization and CSP payoffs.

Esta tese apresenta um conjunto de avaliações experimentais com o objetivo de identificar como as camadas de hardware e software podem ser determinantes na obtenção de desempenho em máquinas virtuais em nuvem. A computação em nuvem apresenta benefícios no fornecimento de serviços de infraestrutura de forma conveniente e sob demanda. Essa oferta de serviços, sob demanda, pode conduzir os clientes à contratação de serviços virtualizados que atendam às suas necessidades com bom desempenho. Quando os serviços de infraestrutura são vendidos ou alugados com a descrição apenas das características das máquinas virtuais, o desempenho obtido pode não ser o desejado devido às restrições dos recursos físicos. Sendo assim, é proposto um levantamento sobre quais informações são fornecidas pelos provedores de nuvem. De posse das informações coletadas, um conjunto de experimentos demonstra a importância da camada de virtualização para os serviços de infraestrutura. Os resultados apresentados nesta tese demonstram que apesar de os provedores terem tipos de máquinas virtuais semelhantes nos recursos virtuais, os recursos físicos são determinantes na obtenção do desempenho. Além dos recursos físicos, os recursos de software também são determinantes na obtenção do desempenho, devido principalmente à marcante influência do virtualizador considerado. / This thesis presents a set of experimental evaluations in order to identify how the hardware and software layers can be crucial in getting services with good performance levels. Cloud computing offers benefits in providing a convenient way of infrastructure services and on demand. This services offer, on demand, can drive customers hiring virtualized services that meet their needs with good performance. When the infrastructure services are sold or leased with a description only of the characteristics of virtual machines, the obtained performance may not be desired due to the restrictions of physical resources. Thus, a survey on what information is provided by cloud providers is proposed. Armed with the information collected, a set of experiments demonstrates the importance of the virtualization layer for infrastructure services. The results presented demonstrate that although providers have similar types of virtual machines in virtual resources, physical resources are determinant in obtaining performance. In addition to the physical resources, the resources of software are also critical in getting the performance, mainly due to the strong influence of the considered virtualizer.

Submitted by João Arthur Martins (joao.arthur@ufpe.br) on 2015-03-11T17:58:21Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao José Lutiano da Silva.pdf: 3688182 bytes, checksum: 3aa20a3af7ab28f4114f4925cef77dcb (MD5) / Approved for entry into archive by Daniella Sodre (daniella.sodre@ufpe.br) on 2015-03-13T13:00:30Z (GMT) No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao José Lutiano da Silva.pdf: 3688182 bytes, checksum: 3aa20a3af7ab28f4114f4925cef77dcb (MD5) / Made available in DSpace on 2015-03-13T13:00:30Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao José Lutiano da Silva.pdf: 3688182 bytes, checksum: 3aa20a3af7ab28f4114f4925cef77dcb (MD5) Previous issue date: 2013-09-19 / Com o advento do uso de serviços hospedados em cenários de computação em nuvem, presencia-se a necessidade de organizações utilizarem seus benefícios. Neste âmbito, empresas governamentais ainda possuem receio em hospedarem seus dados no ambiente de computação em nuvem, o que por sua vez, faz com que muitas dessas entidades utilizem um modelo de oferta de serviços tradicionais não otimizados no contexto de recursos computacionais. Devido a isso, a computação em nuvem privada surge como ambiente alternativo para hospedagem de serviços dentro da própria infraestrutura da empresa pública, suprimindo as preocupações de conformidades à segurança da informação e adequação do modelo de oferta de serviços. Com o objetivo de discutir e ampliar a literatura acerca do assunto são apresentados passos a serem utilizados para implantação de um cenário em computação em nuvem privada, como solução destes problemas em uma empresa pública de processamento de dados. Como contribuição prática, o estudo apresenta uma solução escalável com baixo investimento, uma vez que todo o cenário proposto é utilizado em ferramental de software livre, permitindo impactos positivos e valor agregado no modelo de oferta de serviços do provedor para os clientes.

In mobile cloud computing (MCC), rich mobile application data is processed at the cloud infrastructure by reliving resource limited mobile devices from computationally complex tasks. However, due to the ubiquitous and mobility nature, providing time critical rich applications over remote cloud infrastructure is a challenging task for mobile application service providers. Therefore, according to the literature, close proximity placement of cloud services has been identified as a way to achieve lower end-to-end access delay and thereby provide a higher quality of experience (QoE) for rich mobile application users. However, providing a higher Quality of Service (QoS) with mobility is still a challenge within close proximity clouds. Access delay to a closely placed cloud tends to be increased over time when users move away from the cloud. However, reactive resource relocation mechanism proposed in literature does not provide a comprehensive mechanism to guarantee the QoS and as well as to minimize service provisioning cost for mobile cloud service providers. As a result, using the benefits of SDN and the data plane programmability with logically centralized controllers, a resource allocation framework was proposed for IaaS mobile clouds with regional datacenters. The user mobility problem was analyzed within SDN-enabled wireless networks and addressed the possible service level agreement violations that could occur with inter-regional mobility. The proposed framework is composed of an optimization algorithm to provide seamless cloud service during user mobility. Further a service provisioning cost minimization criteria was considered during an event of resource allocation and inter-regional user mobility.

Focus of this diploma thesis is on Cloud Computing -- a new delivery model of IS/ICT services for companies. Cloud Computing services are IT resources available to users via internet on pay-as-you-use basis. The aim of the thesis is define underlying Cloud Computing concepts and evaluate the current state of Cloud Computing adoption in the Czech Republic. We discuss the framework for definition of Cloud Computing concepts in the theoretical part of the thesis drawing on resources available in the literature. Cloud Computing survey is conducted using an online questionnaire and followsthe principles established in theoretical part. The main contribution of this thesis is creating working definition of Cloud Computing and identifying reasons that affect adoption rate in the Czech Republic.

Esta tese apresenta um conjunto de avaliações experimentais com o objetivo de identificar como as camadas de hardware e software podem ser determinantes na obtenção de desempenho em máquinas virtuais em nuvem. A computação em nuvem apresenta benefícios no fornecimento de serviços de infraestrutura de forma conveniente e sob demanda. Essa oferta de serviços, sob demanda, pode conduzir os clientes à contratação de serviços virtualizados que atendam às suas necessidades com bom desempenho. Quando os serviços de infraestrutura são vendidos ou alugados com a descrição apenas das características das máquinas virtuais, o desempenho obtido pode não ser o desejado devido às restrições dos recursos físicos. Sendo assim, é proposto um levantamento sobre quais informações são fornecidas pelos provedores de nuvem. De posse das informações coletadas, um conjunto de experimentos demonstra a importância da camada de virtualização para os serviços de infraestrutura. Os resultados apresentados nesta tese demonstram que apesar de os provedores terem tipos de máquinas virtuais semelhantes nos recursos virtuais, os recursos físicos são determinantes na obtenção do desempenho. Além dos recursos físicos, os recursos de software também são determinantes na obtenção do desempenho, devido principalmente à marcante influência do virtualizador considerado. / This thesis presents a set of experimental evaluations in order to identify how the hardware and software layers can be crucial in getting services with good performance levels. Cloud computing offers benefits in providing a convenient way of infrastructure services and on demand. This services offer, on demand, can drive customers hiring virtualized services that meet their needs with good performance. When the infrastructure services are sold or leased with a description only of the characteristics of virtual machines, the obtained performance may not be desired due to the restrictions of physical resources. Thus, a survey on what information is provided by cloud providers is proposed. Armed with the information collected, a set of experiments demonstrates the importance of the virtualization layer for infrastructure services. The results presented demonstrate that although providers have similar types of virtual machines in virtual resources, physical resources are determinant in obtaining performance. In addition to the physical resources, the resources of software are also critical in getting the performance, mainly due to the strong influence of the considered virtualizer.

Yes / Failure Prediction has long known to be a challenging problem. With the evolving trend of technology and growing complexity of high-performance cloud data centre infrastructure, focusing on failure becomes very vital particularly when designing systems for the next generation. The traditional runtime fault-tolerance (FT) techniques such as data replication and periodic check-pointing are not very effective to handle the current state of the art emerging computing systems. This has necessitated the urgent need for a robust system with an in-depth understanding of system and component failures as well as the ability to predict accurate potential future system failures. In this paper, we studied data in-production-faults recorded within a five years period from the National Energy Research Scientific computing centre (NERSC). Using the data collected from the Computer Failure Data Repository (CFDR), we developed an effective failure prediction model focusing on high-performance cloud data centre infrastructure. Using the Auto-Regressive Moving Average (ARMA), our model was able to predict potential future failures in the system. Our results also show a failure prediction accuracy of 95%, which is good.

The thesis introduces a study about the vulnerabilities that a company as Scania IT faces towards vendor lock-in in the transition to a cloud computing platform. Cloud computing is a term that refers to anetwork of remote servers hosted on the internet to store, manage and process data, rather than on a local server or a personal computer. Vendor lock-in is an outcome that causes companies to pay a significant cost to move between cloud providers. The effects that cause vendor lock-in that will be described are portability, interoperability and federation are called the lock-in effects. The goal of the research is to help Scania IT understand the vendor lock-in and the vulnerabilities they can face in the transition to the cloud as well as to clarify the concern that they may have against falling in vendor lock-in. The main purpose of the research is to present the various lock-in effects that are related to the transition from one cloud provider to another and the vulnerabilities that cause companies to fall in vendor lock-in. The thesis presents the reasons that motivates why Scania IT would consider using the cloud and the concerns that they may have against usage of a cloud computing platform. The results will be based on a case study of a similar company that has moved to a cloud provider and specifically Microsoft Azure and an interview of Microsoft Azure point of view with the risk of vendor lock-in. Finally, a process of interviews with different people from Scania IT to extract the current bottleneck in the development process that caused the company to think of a cloud computing platform. The results show that companies should consider many risks and factors while moving to the cloud, as vendor lock-in, cloud maturity index and their IT strategies. As a result, the thesis gives recommendations of the steps needed to minimize the risks of the cloud while maintaining the positivity of the cloud. / Uppsatsen presenterar en studie om de sårbarheter som ett företag som Scania IT har mot inlåsning i övergången till molntjänster. Molntjänster är en term som hänvisar till ett nätverk av servrar som finns på internet för att lagra, hantera och processa data, istället för på en lokal server eller en persondator. Inlåsning är ett resultat i vilket orsakar att företagen behöver betala en betydande kostnad för att flytta mellan molnleverantörer. De effekter som orsakar inlåsning vilket kommer att beskrivas är portabilitet, interoperabilitet och federation, dessa kallas inlåsningseffekter. Målet med forskningen är att hjälpa Scania IT att förstå inlåsning och sårbarheter som de kan möta i övergången till molnet. Dessutom är målet att klarlägga riskerna som de kan ha mot att falla i inlåsning. Det huvudsakliga syftet med forskningen är att presentera de olika inlåsningseffekter som är relaterade till övergången från en molnleverantör till en annan samt de sårbarheter som orsakar företagen att falla i inlåsning. Uppsatsen presenterar skäl som motiverar varför Scania IT ska överväga att använda molnet samt den oro som de kan ha mot användning av en molnleverantör. Resultaten kommer att baseras på en fallstudie av ett liknande företag som har flyttat till en molnleverantör och specifikt Microsoft Azure samt en intervju av Microsoft Azure synvinkel med risken för inlåsning. Slutligen, en rad av intervjuer med olika personer från Scania IT för att extrahera den nuvarande flaskhalsen i utvecklingsprocessen som orsakade företaget att tänka på molntjänster. Resultaten visar att företagen bör överväga många risker och faktorer när de flyttar till molnet, som exempelvis inlåsning, cloud maturity index och deras IT-strategier. Som ett resultat ger examensarbetet nödvändiga rekommendationer för att minimera riskerna för molnet samtidigt som positivitet av molnet.

This thesis deals with possible interconnections between two concepts playing a big role in contemporary business and IT world. These concepts are Business Continuity Management and Cloud Computing. In the scope of this thesis there are certain areas identified where both concepts are complement, where Cloud Computing brings new opportunities for Business Continuity Management and where could possible problems arise during particular implementation. From the BCM perspective the impact lies on IT services, from the Cloud Computing perspective the thesis deals especially with security aspects. The thesis is also aimed at the characteristics of higher education and basic differences from commercial sphere. Based on defined differences and identified interconnections between BCM and Cloud Computing, the thesis argues for usage of suitable Cloud Computing solution for higher education regarding Business Continuity improvement. The multi-criterion comparison of several Infrastructure-as-a-Service solutions stems from this analysis focusing on technical, financial, and Business Continuity aspects. The result from this comparison together with conclusions from previous chapters serve as an input for subsequent practical proposal of Cloud Computing solution and its verification against Business Continuity improvement in specific conditions on University of Economics in Prague. The proposal is also represented by strategic map.

Social networking applications that are developed using traditional software and architecture have scalability issues. One way to overcome the high cost of scaling social applications is to use Cloud Computing (CC). There are various cloud computing platforms available. One very interesting CC platform is Google App Engine (GAE). This research focuses on using the free GAE as a way to re-implement existing social networking applications.<p> The research focuses on how to move social applications into the cloud and on the evaluation of their performance. The thesis investigates the GAE platform, and its features. The study shows how to re-implement a social networking application using GAE cloud with limited code approximately 600 lines and evaluates the scalability of the applications.