International Business Machines Corporation

David Frank Mannenbach

Karl Allen Nielsen

Yu-Cheng Hsu

Kevin J Ash

Robert Akira Kubo

James Lee Hafner

David Ray Kahler

Michael T Benhase

Thomas Charles Jarvis

Carl Evan Jones

Kenneth Wayne Todd

Krishnakumar Rao Surugucchi

Richard B Stelmach

Steven Robert Lowe

James A O'Conner

Brent Cameron Beardsley

James A O'Connor

Vernon John Legvold

Said Abdullah Ahmad

Glenn Rowan Wightwick

Tim Stephen Vande Zande

Krishnakumar S Rao

Carol Santich Mellgren

Brian Anthony Rinaldi

Susan Kay Candelaria

Joseph S Hyde, Ii

Robert Louis Morton

Suguang Li

Brent William Yardley

Lawrence Carter Blount

Colleen Renee Stouffer

William Dennis Williams

Basheer Neuhru Bristow

Herve Gilbert Philippe Andre

Walter J Ramsey

A method of managing data in a cache upon a cache write operation includes determining a number of non-contiguously written sectors on a track in the cache and comparing the number with a threshold number. If the number exceeds the threshold number, a full background stage operation is issued to fill the non-contiguously written sectors with unmodified data from a storage medium and the full track is then destaged. A corresponding system includes a cache manager module operating on the storage subsystem. Upon a determination that a cache write operation on a track has taken place, the cache manager module determines a number of non-contiguously written sectors on the track, compares the number with a predetermined threshold number, issues a background stage operation to fill the non-contiguously written sectors with unmodified data from a storage medium if the number exceeds the threshold number, and then destages the full track.

Method to increase performance of non-contiguously written sectors

A method of managing data in a cache memory storage subsystem upon a cache write operation includes determining a first number of non-contiguously written sectors on a track in the cache and comparing the first number with a second, threshold number. If the first number exceeds the second number, a full background stage operation is issued to fill the non-contiguously written sectors with unmodified data from a storage medium. A corresponding system includes a cache manager module operating on the storage subsystem. Upon a determination that a cache write operation on a track has taken place, the cache manager module determines a first number of non-contiguously written sectors on the track, compares the first number with a second, predetermined threshold number, and issues a background stage operation to fill the non-contiguously written sectors with unmodified data from a storage medium if the first number exceeds the second number.

Provided are a method, system, and article of manufacture for error checking addressable blocks in storage. Addressable blocks of data are stored in a storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data for the addressable block. A write request is received to modify data in one of the addressable blocks. The write and updating the checksum are performed in the stripe having the modified addressable block. An indication is made to perform an error checking operation on the stripe for the modified addressable block in response to the write request, wherein the error checking operation reads the data blocks and the checksum in the stripe to determine if the checksum data is accurate. An error handling operation is initiated in response to determining that the checksum data is not accurate.

Error checking addressable blocks in storage

A mechanism is provided for a non-root-level user to execute command line script files via a generic command line interface. A command line interface in the data processing system receives a shellscript command from a non-root-level user via a client system. A security module in the data processing system determines whether the non-root-level user is logged on at a command line interface shell level access. Responsive to the non-root-level user being logged on at the command line interface shell level access, a storage controller at the request of a command line interface identifies a script file associated the shellscript command in a storage of the data processing system. The storage controller executes the script file and the command line interface sends the results of the execution of the script file to the non-root-level user via the client system.

Uploading and executing command line scripts

Provided are a method, system, and article of manufacture for validating stored copies of data images to load into memory. An image of data is maintained in a memory, wherein the image in the memory includes a generation number. The image in the memory is written to at least two copies of the image to storage locations in response to a first event, wherein the generation number for the image in the memory is stored in the storage locations having the copies of the image. A check generation number is stored in a storage location. The image is loaded from at least one of the copies of the image in one of the storage locations to the memory in response to a second event The generation number for the image loaded into the memory. The check generation number is incremented in response to the second event. The generation numbers for the copies of the image in the storage locations and the check generation number are used to validate the copies of the image.

Validating stored copies of data images to load into memory

Provided are a method, system, and article of manufacture for error checking addressable blocks in storage. Addressable blocks of data are stored in a storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block. A write request is received to modify data in one of the addressable blocks. The write and updating the checksum are performed in the stripe having the modified addressable block. An indication is made to perform an error checking operation on the stripe for the modified addressable block in response to the write request, wherein the error checking operation reads the data blocks and the checksum in the stripe to determine if the checksum data is accurate. An error handling operation is initiated in response to determining that the checksum data is not accurate.

Provided are a method, system, and article of manufacture for validating stored copies of data images to load into memory. An image of data is maintained in a memory, wherein the image in the memory includes a generation number. The image in the memory is written to at least two copies of the image to storage locations in response to a first event, wherein the generation number for the image in the memory is stored in the storage locations having the copies of the image. A check generation number is stored in a storage location. The image is loaded from at least one of the copies of the image in one of the storage locations to the memory in response to a second event. The generation number for the image loaded into the memory. The check generation number is incremented in response to the second event. The generation numbers for the copies of the image in the storage locations and the check generation number are used to validate the copies of the image.

Provided are a system and article of manufacture for dumping data in processing systems to a shared storage. A plurality of processing systems receive a signal indicating an event. Each of the processing systems write data used by the processing system to a shared storage device in response to receiving the signal, wherein each processing system writes the data to the shared storage device.

System and article of manufacture for dumping data in processing systems to a shared storage

A RAID system is provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a 'new data, unconfirmed' state. The state of written data is periodically checked, such as with a timer. If the data is in the 'new data, unconfirmed' state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a &#x201c;single dropped write confirmed&#x201d; state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a &#x201c;confirmed good&#x201d; state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an &#x201c;unquantifiable&#x201d; state.

Detection and correction of dropped write errors in a data storage system

An apparatus, system, and method are disclosed for data tracking and, in particular, for facilitating failure management within an electronic data communication system. The apparatus includes a tracking module and an error analysis module. The tracking module stores an adapter identifier in a tracking array. The adapter identifier corresponds to a source adapter from which data is received. The error analysis module determines a source of a data failure in response to recognition of the data failure. The data failure may occur on a host adapter, a device adapter, a communication fabric, a multi-processor, or another communication device. The apparatus, system, and method may be implemented in place of or in addition to hardware-assisted data integrity checking within a data storage system.

Apparatus, system, and method for data tracking

Method, system and computer program product are provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a 'new data, unconfirmed' state. The state of written data is periodically checked, such as with a timer. If the data is in the 'new data, unconfirmed' state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a &#x201c;single dropped write confirmed&#x201d; state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a &#x201c;confirmed good&#x201d; state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an &#x201c;unquantifiable&#x201d; state.

Methods are provided for detecting and correcting dropped writes in a storage system. Data and a checksum are written to a storage device, such as a RAID array. The state of the data is classified as being in a 'new data, unconfirmed' state. The state of written data is periodically checked, such as with a timer. If the data is in the 'new data, unconfirmed' state, it is checked for a dropped write. If a dropped write has occurred, the state of the data is changed to a &#x201c;single dropped write confirmed&#x201d; state and the dropped write error is preferably corrected. If no dropped write is detected, the state is changed to a &#x201c;confirmed good&#x201d; state. If the data was updated through a read-modified-write prior to being checked for a dropped write event, its state is changed to an &#x201c;unquantifiable&#x201d; state.

An apparatus, system, and method are disclosed for copying data from a volatile memory device to a plurality of persistent storage devices in response to a loss of primary power. The apparatus includes a section module, a stripe module, and a write module. The section module sections a data image of a write cache into a plurality of data blocks. The stripe module establishes a plurality of data stripes from the plurality of data blocks. The write module writes in parallel each of the plurality of data stripes to a corresponding plurality of unique, persistent data storage devices in response to a loss of line power to the write cache. Advantageously, the apparatus quickly copies the write data from the write cache to a persistent memory device in a relatively short period of time, decreasing the dependence on significant batteries to back up the volatile memory device.

Apparatus, system, and method for writing stripes in parallel to unique persistent storage devices

Disclosed are a storage controller, and a method of operating a storage controller, for interfacing between host systems and a storage devices system. The storage controller includes a first cluster including a first processor and a first cache, and a second cluster including a second processor and a second cache. The method comprises the step of directing data from the host systems through first and second data paths to the storage system. The first processor and cache are associated with the first data path, and the second processor and cache are associated with the second data path. Under one set of conditions, the controller enters a failover mode, wherein data directed to the first data path are routed to the second data path. Under another set of conditions, the controller deconfigures the first cache without entering the failover mode.

Configuring cache memory from a storage controller

A bad track in a disk device in a data storage system is recovered by determining which cluster in the system the bad track belongs to. If the track belongs to the cluster that discovered that the track was bad, that cluster recovers the track by locating a good copy of the track, staging the good copy of the track to cache, and destaging the good copy of the track storage unit from the cache to all copies of the track, whether good or bad, in the data storage system. If the track belongs to another cluster, the cluster that discovered that the track was bad sends a message to the other cluster to inform it of the bad track, and the other cluster recovers the bad track. In another aspect, all tracks are owned by one server or server cluster, which performs the staging and destaging.

Growing community of inventors

Tucson, AZ, United States of America

David Frank Mannenbach