| 摘要 |
PROBLEM TO BE SOLVED: To enable write omission for a disk to be detected with sufficient accuracy even when the write omission for the disk occurs for a whole part in one block. SOLUTION: A history block (#(N)), to which an update state value indicating the update states, such as a generation, time or a check code, is recorded, is arranged for each of the plurality of blocks (#(0)-#(N-1)) to which the updates are recorded, and these are made as a management unit for a check object. When the update is written onto the disk, the new update state value in the same management unit is calculated, stored in a memory 8 as a confirmation value for the update state, and the new update state value is set up as write data to the history block in the same management unit, the whole management unit including the update and the update state value is written onto the disk. The history block is read afterward, the write omission for the disk is detected by comparing its update state value and the update state confirmation value stored in the memory 8. COPYRIGHT: (C)2006,JPO&NCIPI
|
