International Business Machines Corporation

Steven Douglas Johnson

Edward George Butt

Scott Milton Fry

Steven Bennett Wilson

Vernon John Legvold

Ronald David Martens

Jack Harvey Derenburger

Tony Jing Zhang

David N Mora

Jeffrey E Ferrier

Jed L Dyreng

Matthew G Borlick

Robert Beverley Basham

Alex Chliwnyj

Kirby Grant Dahman

Steven Ross Bentley

Sushama M Paranjape

Franck Excoffier

John James Gniewek

Vincent K Allen

Ronald K Rhodes

A technique for updating elements in a data storage facility, including a single server or a multi-server system, such as by providing updated internal code packages to the elements. The update is performed using a fixed state machine, where the elements are updated in a coordinated manner within the constraints of the state machine. In a multi-server device, code packages are distributed to elements associated with the different servers in one traversal of the state machine, during distribute states of the state machine. The distributed code packages are activated in activate states of the state machine in multiple traversals of the state machine, so there is a serial activation. The code packages can be grouped in a flexible way by configuring an external update bundle used by the state machine. The distributing of the code is based on the grouping.

Updating elements in a data storage facility using a predefined state machine, with serial activation

A method to update code in an information storage and retrieval system comprising one or more processors, and existing code, where that existing code includes a Concurrent Code Load having (N) phases, where that Concurrent Code Load invokes an as yet unwritten Temporal Coupling File. The method provides a code update image comprising that Temporal Coupling File. If the (i)th phase of the Concurrent Code Load invokes the Temporal Coupling File, then the method reads instructions for that (i)th phase of the Concurrent Code Load from the Temporal Coupling File, and executes those instructions. The method completes each phase of the Concurrent Code Load, and then operates the system using the code update.

Apparatus and method to update code in an information storage and retrieval system while that system remains in normal operation

A technique for updating elements in a data storage facility, including a single server or a multi-server system, such as by providing updated internal code packages to the elements. The update is performed using a fixed state machine, where the elements are updated in a coordinated manner within the constraints of the state machine. In a multi-server device, the code packages are distributed to elements associated with the different servers in one traversal of the state machine, during distribute states of the state machine. The distributed code packages are activated in activate states of the state machine in one traversal of the state machine, so there is parallel activation. The code packages can be grouped in a flexible way by configuring an external update bundle used by the state machine. The distributing of the code is based on the grouping.

Updating elements in a data storage facility using a predefined state machine, with parallel activation

A system, method and article of manufacture are provided for the automatic recovery from errors encountered during an automated Licensed Internal Code (LIC) update on a storage controller. The present invention functions with a concurrent or nonconcurrent automated LIC update. The automated recovery from many error conditions is transparent to the attached host system and on-site service personnel, resulting an improvement in the LIC update process.

Automated error recovery of a licensed internal code update on a storage controller

A system, method and article of manufacture are provided for the automatic recovery from errors encountered during an automated Licensed Internal Code (LIC) update on a storage controller. The present invention functions with a concurrent or nonconcurrent automated LIC update. The automated recovery from many error conditions is transparent to the attached host system and on-site service personnel, resulting in an improvement in the LIC update process.

A system, method and article of manufacture are provided to perform an automated concurrent licensed internal code (LIC) update and an automated nonconcurrent LIC update on a storage controller. The automated concurrent LIC update is performed on a storage controller by sequentially updating each cluster while maintaining operation of the storage controller. The automated concurrent LIC update allows multiple hosts to continue to access data via the storage controller while the system is actively undergoing the LIC update. The automated nonconcurrent LIC update is performed during the time the storage controller in not operating. The automated nonconcurrent LIC update allows service personnel to perform the update in an efficient and timely manner. When the update has completed, the storage controller becomes available to process data and the system is operating at the latest LIC level.

Automated control of a licensed internal code update on a storage controller

Optimizing the retrieval of blocks of data from a tape media of a longitudinal serpentine tape drive. A device block map (DBM) is retrieved from the tape media and stored in memory. The DBM comprises a table having one or more rows and one or more columns for each block of data stored on the tape media. The columns are selected from the group comprising a wrap column, a position column, a logical block column, and a file identifier column. One or more retrieval paths for requested blocks of data are determined from an image of the physical layout of the requested blocks on the tape media generated by linear interpolation of the DBM stored in memory. Each of the retrieval paths comprises an ordered sequence of the blocks of data and the manner in which the tape media is to be traversed to accomplish the retrieval of the ordered sequence. A sum of distances is determined for each of the retrieval paths and an optimal one of the retrieval paths is selected based on the determined sum of distances. Thereafter, the blocks of data are retrieved from tape media using the optimal retrieval path.

Host-available device block map for optimized file retrieval from serpentine tape drives

A data storage system for use with a tape cartridge storing data in a serpentine pattern on tape media, and having a memory device for storing a device block map for the tape median, comprises tape drive, or a loading mechanism, which first positions the tape cartridge at memory interface for transfering the device block map from the memory device to a processor, and subsequently positions the tape media for reading and/or writing. The processor processes the device block map from the memory device for optimal retrieval of the serpentine pattern data of the tape media. The processing of the device block map occurs while the tape media is positioned for reading and/or writing, so that the processor begins the retrieval reading process immediately upon the positioning of the tape media. In another aspect of the invention, a robotic accessor of an automated data storage library moves the tape cartridge to the tape drive. The tape cartridge is positioned at the memory interface on the robotic accessor, and the stored device block map is transferred from the memory device to the processor, during the move.

Retrieval of serpentine pattern data using a memory device of a tape cartridge

A computer media tape cartridge has a multiple volume selector switch with a plurality of positions which allow a user to physically segment the tape into a corresponding number of sections for more efficient use of the tape. By moving the switch to a desired position, the user may select and use the tracks in one or all of the positions on the tape. These positions include an 'all position' wherein the tape is a single volume for reading and writing data, and individual positions wherein only the selected position is available for reading or writing data to the tape. Using the cartridge in the latter manner physically segments tape into four separate volumes. The cartridge is used in a tape drive having a plurality of sensors for detecting the position of the switch.

Multiple volume computer media tape cartridge

Various data storage formats help to efficiently locate, read, and write user data stored on magnetic tape media. A tape is formatted by writing multiple segment-headers, free from any interleaved access of user data Adjacent segment-headers are spaced by a predetermined interval to define multiple data storage segments. Segment-headers all contain a unique key, which is copied into a key index to identify valid segments. After formatting, normal tape accesses can be performed. Without erasing any old headers or data, a new formatting scheme can be established by writing new segment-headers on the tape. The new segment-headers include a new unique key, replacing the previous key in the key index. Previous segment-headers stored on the tape are ignored, since they lack the updated key. Segments may be selectively grouped to provide independently addressable partitions. Mapping between segments and partitions can use a fixed relationship (e.g. one-to-one), or each partition may be variably sized according to the amount of data to be stored therein. Variable-sized partitions may be automatically padded with a selected number of empty segments. Another feature is flexible-capacity scaling, which distributes an ordered set of device blocks on a multi-track magnetic tape medium. The device blocks are bi-directionally stored in a continuous configuration of multiple adjacent stacked serpentine patterns occupying some or all of the tape. This configuration permits sequential access of all device blocks without advancing the tape medium to skip over regions between adjacent device blocks.

Tape pre-formatting with uniform data storage segments selectively

A storage media for linear, serpentine recording is pre-formatted for data storage operations by recording partition format information in a single pass along the length of the storage media by recording partition format information for all storage media tracks at regular partition intervals defined by index locations alternately in both storage media directions before moving the storage media to the next partition interval. At each index location, partition markers are recorded in successive parallel tracks first in one media direction and then the other until markers are recorded across the entire width of the storage media. The steps of moving to a next partition index location, recording partition information in successive parallel tracks in first one direction and then the other, and moving to a next index location one partition length away, are repeated until all partitions of the storage media are pre-formatted.

Pre-formatting of a storage media having fixed-size partitions

A method and apparatus for accurately locating targeted append points in a tape drive. The effects of interlayer slip (ILS) are eliminated and smaller IBGs are possible thus providing increased data capacity for a given tape cartridge. A coarse tachometer count is used in combination with the read dataflow to accurately locate the proper append point even with a small IBG size. The appended data are assured to be written using the proper tape velocity which virtually eliminates improper write appends and read temporary errors previously caused by ILS effects. The solution also guarantees that customer data previously written to the tape are not accidentally overwritten.

Growing community of inventors

Tucson, AZ, United States of America

Steven Douglas Johnson