Efficient cross user data deduplication in remote data storage

Cloud data storage service providers such as Mozy, Dropbox, and others perform deduplication to save space by only storing one copy of each file uploaded. While using cloud data storage clients always concern for confidentiality of data. So clients usually encrypt their data, and put it on to the storage. If client specific key is used for encryption then each client has different key so in this case deduplication is not possible. Single key within the network is used for encryption then all data is insecure even if one client is compromised and attacker gets the key. Message-locked encryption resolves this problem. But it is inherently subject to brute-force attacks that can recover files falling into a known set. In DupLESS get and put operations are most expensive operations in terms of time. In proposed architecture, we provide secure deduplicated storage resisting brute-force attacks, and realize it in a modification of existing system DupLESS using blowfish algorithm. Clients encrypt under message-based keys obtained via an oblivious PRF protocol from a key-server. Clients encrypt data and send it to the storage server provider then storage server performs deduplication. It is observed that compare to DupLESS, Proposed System achieve more performance in terms of time. Post process and target deduplication is used at storage server side.