A Framework for Data Protection in Cloud Federations

One of the benefits of cloud computing is that a cloud provider can dynamically scale-up its resource capabilities by forming a cloud federation with other cloud providers. Forming cloud federations requires taking the data privacy and security concerns into account, which is critical in satisfying the Service Level Agreements (SLAs). The nature of privacy and security challenges in clouds requires that cloud providers design data protection mechanisms that work together with their resource management systems. In this paper, we consider the privacy requirements when outsourcing data and computation within a federation of clouds, and propose a framework for minimizing the cost of outsourcing while considering two key data protection restrictions, the trust and disclosure restrictions. We model these restrictions as conflict graphs, and formulate the problem as an integer program. In the absence of computationally tractable optimal algorithms for solving this problem, we design a fast heuristic algorithm. We analyze the performance of our proposed algorithm through extensive experiments.