DISTRIBUTED DATA REDUNDANT STORAGE METHOD BASED ON CONSISTENT HASH ALGORITHM

摘要

A distributed data redundant storage method based on a consistent hash algorithm for the selection of a distribution position of pre-stored data in a storage cluster. The method comprises: first establishing a topologic structure of each storage node in a distributed storage system, and determining the position of each node in a storage node sequence corresponding to each hash subspace using the consistent hash algorithm; then determining the number of copies of data and each copy of storage data in a pre-stored data redundant storage manner; and storing each copy of storage data in a different storage node according to a storage rule.By means of the method, an organization architecture of a cluster in a distributed storage system is rapidly established, and a supervised weight random-selection algorithm is used to achieve the objective of efficiently selecting a storage node when a target storage node sequence is selected for pre-stored data in a cluster, and therefore, the storage requirement of massive data is satisfied.