摘要 |
The embodiment of the present invention provides a method, device and system for finding duplicate data. The embodiment comprises: blocking files to be re-deleted; determining a metadata server responsible for queries through fingerprint information of block data; after finding the fingerprint information of the block data has been already stored in a database, sending the block data which corresponds to the fingerprint information of the block data to a shared file system, without sending to the metadata server for storage. Compared with the prior art, link cost may be greatly reduced and system performance may be improved. |