Abstract: An optimum pathway to data stored on a data storage system having N storage devices and more than N pathways is determined in response to a read request for the data. A sorter separates the read request into an appropriate segment size for sending to the storage devices of the data storage system. An assigner generates the set of read permutations satisfying the read request. A read permutation is selected based on a metric. A collector receives the requested data from the N storage devices in response to the selected read permutation being sent to the storage devices.

Abstract: A data storage subsystem that includes three data storage units, three check storage units, and an array controller coupled to the three data and three check storage units can tolerate failure of any three data and check storage units failures can be occur before data stored on the data storage subsystem is lost. Information is stored on the data storage subsystem as a symmetric Maximum Distance Separation code, such as a Winograd code, a Reed Solomon code, an EVENODD code or a derivative of an EVENODD code. The array controller determines the contents of the check storage units so that any three erasures of the data storage units and the check storage units can be corrected by the array controller. The array controller updates a block of data contained in any one of the data storage units and the check storage units using only six IO operations.

Abstract: Error tolerance is increased for a storage system having a plurality of arrays by making local redundancy in a selected array globally available throughout the storage system. To achieve the increased error tolerance, a donor array is selected from the plurality of arrays when the difference between a minimum distance of the donor array and a minimum distance of a recipient array is greater or equal to 2. A donor storage unit is selected in the donor array and recipient information is then rebuilt from the recipient array on the selected storage unit. The selected storage unit is indicated to the donor array as having been donated before the lost information is rebuilt on the selected storage unit. Preferably, the minimum Hamming distance of the recipient array is d?2 before the donor array is selected from the plurality of arrays.

Abstract: The Hamming distance of an array of storage devices is increased by generating a parity check matrix based on column equations that are formed using an orthogonal parity code and includes a higher-order multiplier that changes each column. The higher order multiplier is selected to generate a finite basic field of a predetermined number of elements. The array has M rows and N columns, such that M is greater than or equal to three and N is greater than or equal to three. Row 1 through row M-2 of the array each have n-p data storage devices and p parity storage devices. Row M-1 of the array has n-(p+1) data storage devices and (p+1) parity storage devices. Lastly, row M of the array has N parity storage devices.

Abstract: A data stream received by a hard disk drive (HDD) is organized into at least one cluster having data blocks and spare data blocks and that is equal in size to an integer multiple of the number of sectors in a track of a disk of the HDD. A variable-index writing technique is used to write each data block of a cluster to a corresponding sector of a track that is encountered by a head when the sector is not defective. A sector of the track is skipped during writing when the sector is defective. The number of spare data blocks in a portion of a cluster corresponding to a single track is reduced by the number of sectors that are skipped in the track so that the number of data blocks plus the number of remaining spare data blocks equals the number of sectors in a track.

Abstract: A data storage subsystem that includes three data storage units, three check storage units, and an array controller coupled to the three data and three check storage units can tolerate failure of any three data and check storage units failures can be occur before data stored on the data storage subsystem is lost. Information is stored on the data storage subsystem as a symmetric Maximum Distance Separation code, such as a Winograd code, a Reed Solomon code, an EVENODD code or a derivative of an EVENODD code. The array controller determines the contents of the check storage units so that any three erasures of the data storage units and the check storage units can be corrected by the array controller. The array controller updates a block of data contained in any one of the data storage units and the check storage units using only six IO operations.

Abstract: An optimum pathway to data stored on a data storage system having N storage devices and more than N pathways is determined in response to a read request for the data. A sorter separates the read request into an appropriate segment size for sending to the storage devices of the data storage system. An assigner generates the set of read permutations satisfying the read request. A read permutation is selected based on a metric. A collector receives the requested data from the N storage devices in response to the selected read permutation being sent to the storage devices.

Abstract: The error tolerance of an array of m storage units is increased by using a technique referred to as “dodging.” A plurality of k stripes are stored across the array of storage units in which each stripe has n+r elements that correspond to a symmetric code having a minimum Hamming distance d=r+1. Each respective element of a stripe is stored on a different storage unit. An element is selected when a difference between a minimum distance of the donor stripe and a minimum distance of a recipient stripe is greater or equal to 2. The selected element is also stored on a storage unit having no elements of the recipient stripe. A lost element of the recipient stripe is then rebuilt on the selected element.

Abstract: Error tolerance is increased for a storage system having a plurality of arrays by making local redundancy in a selected array globally available throughout the storage system. To achieve the increased error tolerance, a donor array is selected from the plurality of arrays when the difference between a minimum distance of the donor array and a minimum distance of a recipient array is greater or equal to 2. A donor storage unit is selected in the donor array and recipient information is then rebuilt from the recipient array on the selected storage unit. The selected storage unit is indicated to the donor array as having been donated before the lost information is rebuilt on the selected storage unit. Preferably, the minimum Hamming distance of the recipient array is d?2 before the donor array is selected from the plurality of arrays.