Rule-Based Snapshot Verification and Recovery Framework for Distributed Cloud Storage

Abstract

Rapid proliferation of cloud storage services has enlarged the issue of protecting user data from  unauthorized modifications, deletions, or constraints by the provider in any distributed system. The current state-of-the-art Cloud-of-Clouds solutions improve both availability and integrity but do not provide an operator-controlled method to verify data across different cloud providers and restore a Metacloud to a previously trusted state after unauthorized modifications. In this paper, we present and implement an Intelligent Snapshot Verification and Recovery Framework (ISVRF) as a web service that integrates four commercial cloud storage providers operating under different legal jurisdictions. The proposed framework uses checksum-based snapshot verification, three synchronization modes (Primary Source, Merge, and Bi-Directional), and incorporates a freeze-and-rollback recovery mechanism. An evaluation of its efficacy is also provided through a snapshot performance workbench tested with different file sizes and varying numbers of providers. In 5 experiments (mean ± SD), the Primary Source strategy consistently achieved the lowest mean latency across all scenarios (e.g., 864.1 ± 60.8 ms for medium files with three providers). The maximum latency increase for the Merge and Bi-Directional strategies was 9.58%. Statistically significant differences between the fastest and second-fastest strategies were found in only 1 out of 9 scenarios (paired t-test, p = 0.0013). Increasing the providers from 3 to 5 resulted in a latency variation of not more than about 7%. These results show that resilient snapshot verification and recovery over diverse jurisdictions can be achieved with relatively modest and primarily transfer-bound overhead, making the framework suitable for periodic data protection in user and organizational contexts where the risk of data loss may be due to factors at the jurisdictional or provider level.