Rating cloud storage service by collaborative remote data checking

Abstract

Cloud storage is a flexible platform for data outsourcing service. In such schemes, users no longer have the physical control of their data. As a result, the reliability of the cloud storage service is the major concern of users. Existing approaches evaluate the cloud storage provider CSP from users' feedback, which are subjective and often biased. In this paper, we propose a new rating algorithm to assess the credibility of CSPs. Our solution is based on technical evidence, e.g., results of the data possession checking, which is a useful approach to ensure the integrity and availability of outsourced data in the cloud. Our approach aggregates the collaborative provable data possession results from the community of clients to achieve high coverage. Meanwhile, we introduce a new reputation-based solution to detect malicious clients based on public verification, and adjust the trust value of such clients. We prototyped our approach and analysed the scheme thoroughly. The results demonstrated that the proposed method is efficient, secure, and has the distinguishing property that supports collaborative verification among composite cloud storage frameworks.