Version-vector based video data online cloud backup in smart campus

As a kind of popular multimedia service, the video business in smart campus is fully developed, and the data volume is experiencing a continuous tremendous growth. However, the video data storage also needs to cope with many security threats resulted from some uncertainly factors. Recently, data backup has been credited as an important method to address these issues. Yet, how to improve the efficiency of video data online backup remains a critical challenge. In the context of cloud computing, a version-vector based video data online backup system is proposed in this paper, which fully implements three important functions, e.g., backup, indexing and recovery. At the time of the client backuping the data to the server, the version vectors of files and data chunks are produced as metadata. Accordingly, only through exchanging a small amount of metadata between the data nodes, it can identify the redundant data such that breaking the limitations on file-level deduplication and data-chunk-level deduplication in the existing backup systems, and can further improve the performance of backup, indexing and recovery. The conducted experiments have been shown to demonstrate the efficiency.