Security-Aware Data Allocation in Multicloud Scenarios

When moving large and heterogeneous data collections to the cloud, a key requirement concerns selection of most suitable (set of) cloud service(s) for outsourcing. Not only can different resources have characteristics requirements, but providers also offer services security guarantees, costs. Selecting single service outsourcing an entire collection result in non-optimal solution, as satisfying, at reasonable costs, all requirements specified by owner might not exist. set could instead ensure satisfaction possibly with economic advantages. In this article, we address problem present flexible expressive, yet simple model supporting owners identifying proper allocation their services. Our allows specify easy intuitive way protection operating granularity level resource (or class thereof), representing minimum guarantees that must store resources. Resources be outsourced plaintext or encrypted form, depending on what is convenient allocation. Data then global apply overall allocation, reduce burden side avoid excessive fragmentation collection. We solve finding satisfies both while minimizing formulating it binary programming problem, thus allowing use existing techniques its efficient solution.