Abstract: A fault-tolerant storage device array using a copyback cache storage unit for temporary storage. When a Write occurs to the RAID system, the data is immediately written to the first available location in the copyback cache storage unit. Upon completion of the Write to the copyback cache storage unit, the host CPU is immediately informed that the Write was successful. Thereafter, further storage unit accesses by the CPU can continue without waiting for an error-correction block update for the data just written. In a first embodiment of the invention, during idle time for relevant storage units of the storage system, an error-correction block is computed for each "pending" data block on the copyback cache storage unit, and the data block and corresponding error-correction block are copied to their proper location in the RAID system.

Abstract: A method for on-line restoration of redundancy information in a redundant array storage system. The invention provides alternative methods of restoring valid data to a storage unit after a Write failure caused by a temporary storage unit fault. In the first preferred method, a valid redundancy block is generated for the corresponding data blocks on all storage units. Resubmitting the interrupted Write operation causes the old (and potentially corrupted) data block to be "subtracted" out of the re-computed redundancy block. The uncorrupted new data block is written over the old data block, and is "added" into the recomputed redundancy block to create a new, corrected redundancy block. The new, corrected redundancy block is written to the appropriate storage unit. In the second preferred method, a new redundancy block is generated from all valid data blocks and the new data block. The new redundancy block and the new data block are then written to the appropriate storage units.

Abstract: A method for on-line reconstruction of a failed storage unit in a redundant array system. After providing a replacement storage unit for the failed storage unit, reconstruction begins for each data stripe in the array. General reconstruction consists of applying an error-correction operation (such as an XOR operation on data blocks and a corresponding parity block) to the data blocks from the remaining storage units in the redundancy group, and storing the result in the corresponding block of the replacement storage unit. If a Read operation is requested by the CPU for a data block on the replacement storage unit, then a concurrent Read task is executed which reconstructs the stripe containing the requested data block. If a Read operation is requested by the CPU for a data block not on the replacement storage unit, a concurrent Read task is executed which performs a normal Read.

Abstract: An arrayed disk drive system for providing memory to a computer, said arrayed system having a plurality of disk drives configured to form an array, said arrayed disk drives accessed by a plurality of channels, each channel accessing a plurality of disk drives, including a means for controlling the logical configuration of the arrayed disk drives to appear to the computer as any conceivable arrangement of disk drives, whereby the arrayed disk drive may appear to the computer as the plurality of disk drives, or as one large disk drive comprised of all the arrayed disk drives, or any combination in between. A means for providing a plurality of levels of redundancy on data read or written by the computer to the arrayed disk drives is provided, as well as means controlled by the controlling means for enabling from none to the plurality of levels of redundancy to operate on data being read or written from the arrayed disk drives.