International Business Machines Corporation

Simon Walsh

John Paul Agombar

Carlos Francisco Fuente

Christopher B Beeken

Michael Huw Francis

Robert Bruce Nicholson

Christopher B E Beeken

Paul Andrew Ashmore

Stephanie Machleidt

Christopher Barry Edgar Beeken

Matthew John Fairhurst

Barry Douglas Whyte

Colin R Jewell

Huw Francis

Geoff Lane

Barry John Wood

A mechanism is provided in a computing system for controlling virtualized storage operable to communicate with a host and with mapped and unmapped storage resource pools. A selection component selects a target for a destructive data storage operation from the mapped storage resource pool. Responsive to the selection of the target, a virtual targeting component creates a virtual target from the unmapped storage resource pool to represent the target. Responsive to the selection of the target, a storage move component moves the target to a protected storage resource pool. Responsive to the creation of the virtual target from the unmapped storage resource pool, storage move component, moves the virtual target to the used storage resource pool. The computing system then performs the destructive data storage operation on the virtual target.

Stored data protection and recovery

The present invention provides for a method, system, and computer program product for managing the storage of data. Data is selectively compressed based on a pre-defined compression policy and metadata is stored for physical storage blocks. A stored compression policy identifies at least one criterion for compression, and physical blocks of data meeting the compression policy are identified. A physical block is selected as a source block for data compression, and one or more physical locations are selected as target locations. Data is read from the source block, compressed, and written to the target locations. Metadata is updated to indicate a mapping between the target locations and the virtual blocks previously mapped to the source block. Extra storage capacity can be freed up until more physical storage is ordered and installed, while more important data, such as recently or frequently accessed data, is retained in an uncompressed and accessible state.

Method, system and computer program product for managing the storage of data

Managing resource reclamation in data storage systems is provided. A volume deletion metadata recorder records metadata for one or more deleted volumes. A policy engine, responsive to a predetermined policy rule, applies the policy rule to the metadata. The policy engine initiates policy-controlled data storage space reclamation for the one or more deleted volumes. A volume reclaimer, responsive to the policy engine, reclaims data storage space from the one or more deleted volumes. A resource allocator allocates the data storage space to satisfy a minimum requirement for available zeroed extents that comprise a minimum requirement to satisfy needs of late allocated storage volumes.

Resource reclamation in data storage

The present invention provides for a method for managing the storage of data in a computing system that includes a data processor and local physical storage, involving the steps of: defining a virtual storage volume for access by the data processor, the data processor including a local storage pool mapped to the local physical storage and a remote storage pool mapped to physical storage at a remote site, and the virtual storage volume being overallocated with respect to the local storage pool; and migrating data between the local storage pool and the remote storage pool according to a defined migration policy.

Data migration

The present invention relates to methods of copying and updating data in a processor memory during program run-time without suspending the program's access to its data in the memory during the data-copying and updating processes. In one aspect, the invention concerns a method of copying data from processor memory to a dump memory during run-time, including determining whether data in a segment of the processor memory is to be updated by a program and copying the data to a corresponding segment in the dump memory prior to the data being updated. According to one embodiment, a data-copying instance is utilized to obtain a point-in-time image of a data content of processor memory while the program is running.

Method and system for copying the data contents of a computer system memory during application program run-time

An apparatus, method, and computer program for facilitating disaster recovery of a first computer system, wherein first data residing on a first storage device associated with the first computer system is recoverable from second data residing on a second storage device associated with a second computer system and wherein the second storage device is operable to have an associated state of powered up or powered down. A receiver receives an update operation for updating the second storage device. A processor updates a non-volatile storage means with the update operation, prior to update of the second storage device, wherein the non-volatile storage means is associated with the second computer system.

Apparatus for facilitating disaster recovery

A storage controller includes an interface to host computer apparatus and an interface to a plurality of controlled storage apparatus. The storage controller comprises a host write component operable to send a request to write a data object to a source data image at one of said plurality of controlled storage apparatus; a copy component operable in response to a metadata state to control copying of said data object to a target data image in a delimited sequence of data images in a cascade at said plurality of controlled storage apparatus and a fill-in component operable in response to detection of an offline condition of a disk containing said target data image to control substitution of a fill-in disk in place of said disk containing said target data image.

Storage system with cascaded copy targeting and enhanced integrity

An apparatus is disclosed in which a storage controller cooperable with a host and a plurality of controlled storage is provided to localize an impact of a failure to a target disk in an affected segment. The storage controller includes a host write component to write a data object to a source image storage; a first copy component responsive to a first metadata state to control copying of the data object to a first target storage; a second copy component responsive to a second metadata state to perform either: copying the data object to a second target or causing the first copy component to copy the second target to the first target; and a third copy component to control cascaded copying of the data object to a third target storage. Either the second or the third copy component controls cascaded copying of a delimited data image subsequence responsive to a metadata state indicating currency of a data grain in either the second or the third target.

Storage system with multiple copy targeting and disk failure protection

A method and apparatus is disclosed in which a storage controller cooperable with a host and a plurality of controlled storage is provided to localize an impact of a failure to a target disk in an affected segment. The storage controller includes a host write component to write a data object to a source image storage; a first copy component responsive to a first metadata state to control copying of the data object to a first target storage; a second copy component responsive to a second metadata state to perform either: copying the data object to a second target or causing the first copy component to copy the second target to the first target; and a third copy component to control cascaded copying of the data object to a third target storage. Either the second or the third copy component controls cascaded copying of a delimited data image subsequence responsive to a metadata state indicating currency of a data grain in either the second or the third target.

A method for maintaining data access during failure of a controller in a multiple controller storage subsystem () is provided. The storage subsystem () has an array of data storage devices () and more than one controller () for managing the data storage. The method comprises a first controller () saving its internal state information () and, optionally, resetting itself (). One or more of the other controllers () carry out the steps of pausing operation of the controller (), saving internal state information of the controller at the time of pausing (), and continuing operation of the controller (). The one or more other controllers () may pause operation and save their internal state information when they receive a message broadcast () from the first controller () which has detected an error.

Maintaining data access during failure of a controller

A method for data retention in a data cache and a data storage system are provided. The data storage system () includes a storage controller () with a cache () and a data storage means (). The cache () has a first least recently used list () for referencing dirty data which is stored in the cache (), and a second least recently used list () for clean data in the cache (). Dirty data is destaged from the cache () when it reaches the tail of the first least recently used list () and clean data is purged from the cache () when it reaches the tail of the second least recently used list ().

Method for data retention in a data cache and data storage system

The invention relates to caching data in a data processing system including a host computer and a storage subsystem including at least one customer disk and a cache disk. Write transactions are received specifying data to be written to at least one customer disk and these are cached in a volatile memory of the storage subsystem and written to the cache disk. The transaction data is written sequentially to the cache disk when available. In the intervening periods (i.e., when no transaction data is available), padding data is instead written sequentially to the cache disk.

Growing community of inventors

Hayling Island, United Kingdom

Simon Walsh