Audinet: A Decentralized Auditing System for Cloud Storage

Meng Yan, Jiajia Xu, Trent G. Marbach, Haitao Li, Gang Wang, Xiaoguang Liu

Abstract: In cloud storage, remote data auditing is designed to verify the integrity of cloud data on behalf of cloud users. The audit is performed by a third-party auditor (TPA) according to auditing protocols such as proof of retrievability and provable data possession. However, the TPA-based auditing framework leads to single-point failures, opaque audit processes and undetected mistakes. In this paper, we propose a decentralized auditing system in which the audit is performed by multiple auditors and the audit result is reached in a collaborative and transparent way. Auditors are selected for each audit randomly from the set of cloud users via modified cryptographic sortition; auditing procedures are implemented using a smart contract, and auditing records are published on a blockchain; an incentive mechanism is provided to regulate the behavior of system participants. We implement a prototype system and demonstrate that the proposed system is reliable and technically feasible.