Secure deduplication for big data with efficient dynamic ownership updates

Abstract

With the explosive growth of data on the internet, the trend is for users to store these large amounts of data on the cloud. To protect the privacy of data, users need to encrypt their data before uploading them to the cloud. This brings a big challenge for cloud to perform deduplication. In order to tackle this challenge, we design a secure deduplication scheme for big data with efficient dynamic ownership updates. The proposed scheme can efficiently support user enrollment and revocation. We adopt a new locating technique to effectively locate the re-encrypted data block. As a result, only a small proportion of data needs to be re-encrypted. When one user deletes or updates one file, this user will lost the access to the original data. We also provide the security analysis and experimental evaluation results for the proposed scheme.