Heterogeneous deployment of cloud applications
The different proposals regarding automated cloud deployment respond to the need to deal with supplier heterogeneity and the different levels of service, which requires deal with portability and interoperability problems. In the context of the Cloud, there are two levels of heterogeneity: on one hand, the differences imposed by the features of the business and the infrastructures of the different suppliers, and on the other, the difference in the levels of the service offered (basically, IaaS or PaaS).
This line of research proposes deployment methodologies which can generate in a (semi-) automatic manner, on the basis of an application’s architectural definition (components, relation between these and specification of restrictions), the deployment over different suppliers and at different levels of service. The line of research also considers the capacity to dynamically modify the application’s topology, in such a way that the migration of components between suppliers and between levels of service is also feasible, and can be performed at runtime.