Privacy-preserving layer over MapReduce on cloud

Cloud computing provides powerful and economical infrastructural resources for cloud users to handle ever-increasing Big Data with data-processing frameworks such as MapReduce. Based on cloud computing, the MapReduce framework has been widely adopted to process huge-volume data sets by various companies and organizations due to its salient features. Nevertheless, privacy concerns in MapReduce are aggravated because the privacy-sensitive information scattered among various data sets can be recovered with more ease when data and computational power are considerably abundant. Existing approaches employ techniques like access control or encryption to protect privacy in data processed by MapReduce. However, such techniques fail to preserve data privacy cost-effectively in some common scenarios where data are processed for data analytics, mining and sharing on cloud. As such, we propose a flexible, scalable, dynamical and cost-effective privacy-preserving layer over the MapReduce framework in this paper. The layer ensures data privacy preservation and data utility under the given privacy requirements before data are further processed by subsequent MapReduce tasks. A corresponding prototype system is developed for the privacy-preserving layer as well.