| 摘要 |
PROBLEM TO BE SOLVED: To reduce a data area or period such that the number of redundancy is reduced when a cluster member withdraws.SOLUTION: In a cluster system, a dispatcher 202 includes: a storage part for storing information indicating an area in charge of data on an ID space for each processor; and a storage information management part for, immediately when the processor configuring a cluster withdraws, recombining the area in charge of data on the ID space by referring to the stored information indicating the area in charge to reconfigure the cluster, and for notifying any dispatcher other than the dispatcher itself of the change. A processor 203 includes: an area-in-charge update recording part for, each time the cluster is reconfigured, recording the information indicating the area in charge of data on the ID space in a local storage; and a data update recording part for, each time the data stored in the local storage is updated, recording information indicating that the data have been updated and the information indicating the area in charge in the local storage. |
