RDIMM: Revocable and dynamic identity-based multi-copy data auditing for multi-cloud storage

Abstract

Multi-copy data possession on multi-cloud storage service provides resource-constrained users with a secure and effective way to process data maintenance. To enable integrity for the outsourced data, numerous multi-copy data auditing schemes for multi-cloud storage have been proposed. Nevertheless, existing solutions suffer from the following limitations: (i) cannot support dynamic data updates; (ii) fail to consider the user revocation feature; (iii) may leak the privacy of user identity. To address these limitations, we propose a new identity-based multi-copy data auditing scheme for multi-cloud storage, termed RDIMM. We introduce a new variant of Merkle Tree to enable fully dynamic data updates, in which all copies of a raw block are kept in the same leaf node. In addition, we design a new key generation strategy and a private key update technique, where the cost of user revocation is independent of the total number of file blocks owned by the revoked user. Moreover, we conduct a comprehensive correctness analysis and security analysis of our proposed RDIMM. In addition, we provide detailed theoretical analysis and experimental simulations, which illustrate the effective functionality and practical performance compared to state-of-the-art works.