ARQDEP: arquitetura de computação em nuvem com dependabilidade

• Ano de defesa: 2014
• Autor(a) principal: Lima, Geycy Dyany Oliveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Ciência da Computação; Ciências Exatas e da Terra; UFU
• Palavras-chave em Português: Computação em nuvem; Dependabilidade; Serviços; Confiabilidade; Disponibilidade; Cloud computing; Dependability; Service; Reliability; Availability; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufu.br/handle/123456789/12557; https://doi.org/10.14393/ufu.di.2014.19

Cloud Computing has emerged as one of the most influential technologies of the IT industry, and is presently revolutionizing the way by which IT resources are used and managed. To test dependability environments becomes something which is both important as well as difficult for these environments, in virtue of the infrastructures heterogeneity. Cloud Computing is treated as a technology that requires a high level of trust between the provider and the client. The definitions concerning Cloud Computing are discussed for a better understanding of the proposal made in this Master s Dissertation. The essential features of Cloud Computing are defined in a clear and brief manner; the service models recognized by NIST, the cloud implementation models, the SLAs along with two provider architectures currently used on the market are also discussed. The concepts behind dependability are introduced, as also are the threats and the means by which dependability can be reached. A dependability based Cloud Computing architecture denominated ArqDep is presented. The ArqDep architecture is implemented into an Infrastructure as a Service - IaaS, in the private Cloud implementation model. To aid in the architecture s implementation a Cloud management platform called OpenStack was used. An infrastructure management model for Cloud Computing is proposed: this model carries out the monitoring and control of the infrastructure. A system failure tree is elaborated, which is based upon the failure classification model. Computer resource metrics that affect availability and confidence in the system are collected from the infrastructure and mechanisms responsible for the relocation of resources are used.