International Business Machines Corporation
Method and apparatus for data deduplication

Abstract:

A current file is obtained in the data. It is determined whether a similar historical file exists based on a sampled data block from at least one predetermined location in the current file. In response to non-existence of the similar historical file, the current file and corresponding metadata are stored on a file basis. In response to existence of the similar historical file, a deduplication operation is applied on the current file on a block basis.