| 摘要 |
Embodiments of the present invention provide a data processing method, a system and a client. A target storage node is determined in a manner of comparing a second vector that receives data with a first vector that corresponds to all storage nodes and is pre-stored on a client that receives the data. The target storage node no longer needs to be determined in a manner of sampling some fingerprint values from the data that is received, sending them to all the storage nodes in a data processing system for query and waiting for feedback from the storage nodes, thus preventing multiple times of interaction between the client and the storage nodes, improving deduplication performance, reducing network bandwidth occupation and reducing a delay in a meanwhile. |
