Secure Deduplication on Public Cloud Storage
Hendrik Graupner, Kennedy Aondona Torkura, Muhammad I.H. Sukmana, Christoph Meinel · 2019
Public cloud storage is an important resource of modern computing. One important aspect is the potential to improve economic efficiency. It can be drastically increased by leveraging deduplication techniques. The authors of this paper aim to minimize the risk of data confidentiality violation while maximizing storage utilization. This secure and efficient deduplication concept combines the benefits of context-sensitive partitioning, convergent encryption and remote update protocols. Those technologies are enhanced by a partition-based sensitivity detection algorithm and the handling of sensitive partitions. Thus, incremental updating of stored files is possible. The scheme features an information concealment algorithm to prevent information retrieval attacks. In addition, our solution prevents gaining of illegitimate access to sensitive data due protection by the proof-of-data-ownership scheme Asymmetric Convergent Encryption (ACE). This holistic concept overcomes the weaknesses of deduplicating storage systems. The results of this paper can be leveraged to enhance not only public cloud storage but also other remote storage systems.