A UTILIZAÇÃO DO POSICIONAMENTO DE MÁQUINAS VIRTUAIS PARA A QUALIDADE DE APLICAÇÕES DE CLIENTES EM NUVEM

Abstract

With the evolution of virtualization technologies, networks and computing hypervisors, resource sharing in today's infrastructures as services has become more flexible. The entire implementation of a cloud depends on performance and standardization for customers; Given that the environment is shared, everyone wants to get the best performance. Service qualities are defined, protocols are standardized, and there is concern for security among customers, where resource quotas are created and, in theory, the platform is ready for demand. But performance for customers varies, and resource degradation soon becomes a barrier to the efficient use of cloud infrastructure. The compromise of scalability and availability is due to the lack of a strategy for hardware heterogeneity, variations of network factors, computing and workloads, which are the characteristics of a shared environment, requiring a technology capable of satisfying Automation of the rules proposed by the cloud model. These characteristics are studied in this article, as well as the key concepts of cloud computing, virtual network, data center models and placement algorithms.