| 摘要 |
A method and apparatus for improving the data storage reliability of a host computer writing to a disk storage device receive a plurality of blocks of data from the host computer and write the blocks of data, according to a selected format, in a high speed cache memory. The data blocks in cache memory are then written to the disk drive, the blocks having a defined size. A plurality of the blocks form a sector on the disk drive and a plurality of the sectors form a track on the disk drive. Writing to the disk includes, transparently to the host, the calculation and attachment to each block of an error correcting code value. When data is written to the data cache, there is generated and associated with each of a group of the blocks of data, from the host computer, a second error code. The second error code is stored at a physical location in the cache memory, and when written to disk, is further written at a location which is not in the same sector as any of the blocks with which the second codes are associated. In this way, the host to cache level communications acquires a higher degree of reliability than previously available for full block transfer under, for example, the UNIX operating system. |
