Abstract: In accordance with data storage and recovery at a spatial failure pattern, a storage and recovery algorithm based on data replication was proposed. The algorithm assigns a redundant node for every data node, and it can recover data successfully from the redundant node when the data nodes fail and thus minimise the total cost. Compared with the Greedy algorithm, theoretical analysis and simulation results show that this method can not only reduce the total energy cost but also prolong the lifetime of networks for successful recovery.

Key words: data storage, node failure, data replication, data recovery