Rebuilding lost data in a distributed redundancy data storage system

摘要: Rebuilding lost data in a distributed redundancy storage system including multiple nodes, is provided. User stored as collection of stripes, each stripe comprising strips and associated parity strips, the stripes across corresponding owner nodes nodes. A node maintains strip holding first copy data, for strips. Upon detecting failure condition, initiates rebuilding protocol recovery and/or it owns. The includes reconstruction or by computation involving from set stripe, wherein contains at least one surviving strip.