Abstract: A method of taking a backup copy of a data set, for use with a log, with the data set being open for update by multiple updaters is described. A point on the log is identified (Recovery Time) from which it can be ensured that all data updates identified on the log have been written by an updater from buffers to the data set. This is done by recording a Recovery Time, and then not taking the backup until all the updaters have confirmed that they have written all updates from buffers to the data set. In addition, if tokens are used on the log to avoid having to identify the data set in full, then these are also written to the log before the update is taken.

Abstract: In a storage system comprising an array of storage devices including a processor and memory, a plurality of data blocks are organized on the storage devices in segments striped across the storage devices. A main directory stored in memory contains the location on the storage device of each of the current data blocks. A method is provided for recovering a main directory in the event that the main directory becomes damaged. A segment directory having information on the data blocks in the segment is stored as part of the segment on the storage devices. Periodically a checkpoint of the main directory is written to the storage devices. A list is maintained of closed segments written to the storage devices since the checkpoint directory was written to the storage devices.

Abstract: If consecutive read or write requests imposed on a DASD are of the same type and bear a defined sequential logical address relationship (pure sequential, near sequential), then a circular buffered data path using a pair of a synchronously managed read/write ports respectively coupling either a cyclic, concentric, multitracked storage medium or a cyclic, spiral-tracked storage medium and a device interface can continue data streaming unabated. Otherwise, the path would ordinarily have to be disabled and reconnected using a control microprocessor in respect of any random sequence of requests.

Abstract: A data processing system comprises a host computer connected for the transfer of data to and from a plurality of data storage devices arranged in a string, the host computer including communication means comprising first and second ports connecting to first and second communication links, the first and second communication links being connected respectively to first and second data storage devices of said string. A bypassing means is provided between the first and second ports of the host system and the first and second data storage devices, the bypassing means being comprised of an independent bypass circuit on each of the first and second communication links between each of the first and second ports and the first and second data storage devices, the bypassing means being operable to bypass the host computer by connecting the first and second devices only when both of said independent bypass circuits detect a lack of data transfer on their respective links.

Abstract: In a disk array system, such as a RAID system, when disks are added thereto, this invention rearranges data without requiring a back-up of the stored data. When new data storage units are added into a RAID system in which each data storage unit is a row and an address range of the data storage unit is a column, a new striping rule dependant on the number of data storage units in the system, inclusive of newly added unit, is determined. Data are rearranged in the storage areas according to the new striping rule. In rearranging one row according to the new striping rule, the data involved in the transfer to one row from the other row, continues to be stored as a place holder block. Data that has been rearranged is accessed according to the new striping rule which data that has not been rearranged is accessed according to the old striping rule.

Abstract: A method for rebuilding contents of a malfunctioned direct access storage device within a log-structured array is disclosed. In accordance with the method and system of the present invention, each direct access storage device within a log-structured array is divided into multiple segment-columns, and each corresponding segment-column from each direct access storage device within the log-structured array forms a segment. A segment is first located within the direct access storage devices. A determination is made as to whether or not the segment is empty. In response to a determination that the segment is empty, a pointer is moved within a segment-column mapping table from pointing to a segment-column in the malfunctioned direct access storage device to point to a segment-column in a spare direct access storage device of the segment.

Abstract: An actuator arm has a recess portion in its side surface which is positioned vertically during manufacture. A suspension having etched passages is placed next to the actuator arm such that the passages intersect the recess portion. A liquid adhesive is placed in the recess portion which act as an adhesive reservoir. The adhesive is then drawn into the passages and is cured such that the suspension is bonded to the actuator arm.

Abstract: Disclosed is an array controller for controlling the transfer of data from a host system to an array of data storage devices, comprising a processor connected via a local bus to a data buffer in which data is staged during said transfer. The array controller is provided with a buffer controller for controlling the operation of the buffer and is further provided with channel hardware for manifesting a plurality of data channels, selectable by the local bus address, over which data is transferred in and out of the data buffer.

Abstract: This invention provides a one-pass storage process to manage storage space in a storage hierarchy system wherein whole objects or fragments of whole objects can be retrieved efficiently. Metadata to represent appropriate geometric characteristics of storage devices, units of transfer to minimize retrieval time, and buffers, are used to control the storage allocation. A plurality of objects from an object set are aggregated into at least one object grouping for storage where the object grouping is smaller than a minimum storage size. For each object grouping and each object not in an object grouping, a determination is made of the total number of blocks of storage needed and a minimum transfer size. Extents of blocks are identified in the storage device of a size greater than the minimum transfer size totalling the total number of blocks of storage needed.

Abstract: A method and system are provided for improving the speed for writing data in a disk array system (RAID). Disks are divided into the ones for reading old data and the other ones for writing new data. An external semiconductor storage device is used as a parity device which significantly enhances performance in writing.A disk array system is used for storing data for which a parity group comprises N data units stored in sectors of N magnetic disk drives with parity stored in the parity device. At least N+1 data-storing magnetic disk drives are provided, where sectors of at least one magnetic disk drive are used as redundant sectors.

Abstract: A system and method are provided for managing persistent and non-persistent data in a log structured disk array. In a log structured array, data are first collected in a log stored in a buffer. When the buffer is full, the data are written to a disk drive. Persistent data are continually accessed over a long period of time such that, most disk activity is associated with persistent data. Preferably, the persistent data are stored in the part of the disk where access times are minimal (active region), such as in the middle cylinders of a disk. When writing data (including during garbage collection procedure) persistent data is written to the active region and non-persistent data is written to the other regions (inactive region). The persistence of the data is monitored. Persistent data is migrated to the active region and non-persistent is migrated to the inactive region.

Abstract: A method and system for automatically allocating space within a data storage system for multiple data sets which may include units of data, databases, files or objects. Each data set preferably includes a group of associated preference/requirement parameters which are arranged in a hierarchical order and then compared to corresponding data storage system characteristics for available devices. The data set preference/requirement parameters may include performance, size, availability, location, portability, share status and other attributes which affect data storage system selection. Data storage systems may include solid-state memory, disk drives, tape drives, and other peripheral storage systems. Data storage system characteristics may thus represent available space, cache, performance, portability, volatility, location, cost, fragmentation, and other characteristics which address user needs.

Abstract: A system is provided for storing data for a plurality of host computers on a plurality of storage arrays so that data on each storage array can be accessed by any host computer. A plurality of adapter cards are used. Each adapter has controller functions for a designated storage array. There is an adapter communication interface (interconnect) between all of the adapters in the system. There is also a host application interface between an application program running in the host computer and an adapter. When a data request is made by an application program to a first adapter through a host application interface for data that is stored in a storage array not primarily controlled by the first adapter, the data request is communicated through the adapter communication interface to the adapter primarily controlling the storage array in which the requested data is stored.

Abstract: A method for rebuilding contents of a malfunctioned direct access storage device within a log-structured array is disclosed. In accordance with the method and system of the present invention, each direct access storage device within a log-structured array is divided into multiple segment-columns, and each corresponding segment-column from each direct access storage device within the log-structured array forms a segment. A segment is first located within the direct access storage devices. A determination is made as to whether or not the segment is empty. In response to a determination that the segment is empty, a pointer is moved within a segment-column mapping table from pointing to a segment-column in the malfunctioned direct access storage device to point to a segment-column in a spare direct access storage device of the segment.

Abstract: A method and system for managing the movement of large video data files from an archival tape storage to an active hard-disk storage within a video server computer system are disclosed. In accordance with the method and system of the present invention, a storage list, a lookahead list, and a candidate list are produced. Each item on the storage list represents a multimedia data file stored within an active storage medium. Each item on the lookahead list represents a multimedia data file that will be played at a date and time according to a play schedule. Each item on the candidate list represents a candidate multimedia data file for removal from the active storage medium because it has already been played. After a simulation of a multiple of events utilizing the storage list, the lookahead list, and the candidate list, an output schedule file based on the results from the simulation is developed.

Abstract: An object of the present invention is to provide a data storage system and a parity generation method for the data storage system which make generation of parity easier and which are capable of performing the reconstruction of data at the time of an occurrence of failure with high efficiency and high speed.A data storage system is provided comprising a plurality of devices for storing data, a plurality of data buses for transferring the data, and a selector connected between the plurality of devices and the plurality of data buses for selectively connecting a device and a data bus using a predetermined combination, the selector including a parity operation generator.

Abstract: When an update is made to a data record in a distributed, replicated data processing system, the update is first applied to a primary copy of the data record before being applied to any other copy to ensure that updates are applied in the correct time sequence. Apparatus and a method of operating a distributed data processing system is provided in which responsibility for the primary copy is transferable to whichever processor in the system requires most frequent update access, providing improved performance and availability of data. The primary copy may be partitioned and distributed across the system.

Abstract: A method for correcting error bursts in data which are uncorrectable by an ECC in a communications channel in which no pointer is available or if available is not generated. The data is recorded in blocks preferably comprising subblocks. Parity cell syndromes are generated for each block during writing; and during reading these syndromes are analyzed to identify possible starting points of error burst locations. A trial correction is applied to the data in the uncorrectable block (or subblock), then verified whether successful using CRC. The correction is accepted as valid if and only if only one trial correction is verified as successful. The maximum number B of consecutive bytes in error correctable in a block is less than, the number N of bytes in each parity cell, and N-B is selected to limit the probability of miscorrection.

Abstract: A method is described for operating a computer to sort a set of data records each having an associated key for governing the sort process, the method comprising determining a range for the key values by sampling the key values; defining a plurality of buckets, each bucket corresponding to a respective one of a plurality M of subintervals in the range, and two edge buckets for key values outside the range, each subinterval having a respective index; distributing the keys among the buckets by determining directly from each key value the index of the subinterval into which the key value falls; and processing the buckets in sequence in order to sort the records, sorting the keys in each bucket if the bucket contains more than one key.

Abstract: A system and method for changing the number of logical volumes in a drawer in a rack in a direct access storage device subsystem is disclosed. The method and system are able to change the number of logical volumes without disrupting access to the other logical volumes in the rack. Channel connection addresses, which are logical volume addresses as known by the CPUs, are freed by removing the old drawer and then are mused. If the new drawer has more logical volumes than the old drawer, the next unused channel connection addresses are used with the new drawer. In a subsystem having a storage controller for providing control for a plurality of direct access storage devices, the logical volumes are spread across multiple physical devices. The storage controller maintains configuration data for the entire subsystem in redundant, non-volatile storage locations reserved specifically for its use.