International Business Machines Corporation

Mark Albert Reid

Joseph M Swingler

Thomas W Bish

Jonathan Wayne Peake

Brent Cameron Beardsley

Susan Kay Candelaria

Dean Lee Hanson

Michael T Benhase

Gregory Tad Kishi

Robert Louis Morton

Thirumale N Niranjan

Douglas A Martin

Michael Wayne Young

Kenneth Wayne Lane

Kenneth Fairclough Day, Iii

Douglas William Dewey

Michael Howard Hartung

James Arthur Fisher

Vernon John Legvold

John Norbert McCauley

Keith Anthony Bello

Donald M Nordahl

William G Sherman, Ii

Bradley S Powers

Carol Santich Michod

Robert L Kwok

Dennis Albert Kukula

Guy Eugene Martin

Jonathan Wayner Peake

William Lee Richardson

A data storage system stores logical data object(s), each identified by a logical identifier. A control is configured to assign a unique WORM (Write Once Read Many) identifier to the logical data object, and stores the unique WORM identifier as associated with the logical identifier, in a database maintained by the control so as to be persistent. Data storage is configured to write the logical data object with a header with the unique WORM identifier. The control, in order to allow the logical data object to be accessed externally to the control, requires matching the unique WORM identifier in the header of a logical data object to the unique WORM identifier of the persistent database for the logical object. The unique WORM identifier is formed of a checksum hash value related to nonce fields comprising at least the logical identifier of the logical data object, and an incrementing token.

Extended logical worm data integrity protection with unique worm identifier in header and database

A data storage system stores logical data object(s), each identified by a logical identifier. A control is configured to assign a unique WORM (Write Once Read Many) identifier to the logical data object, and stores the unique WORM identifier as associated with the logical identifier, in a database maintained by the control so as to be persistent. Data storage is configured to write the logical data object with a header with the unique WORM identifier. The control, in order to allow the logical data object to be accessed externally to the control, requires matching the unique WORM identifier in the header of a logical data object to the unique WORM identifier of the persistent database for the logical object. The unique WORM identifier is formed of a checksum hash value related to nonce fields comprising at least the logical identifier of the logical data object, an incrementing token, and a time stamp.

Local versions of attributes of a storage object are maintained at a plurality of nodes, wherein a first attribute designates a first node of the plurality of nodes as an owner node for the storage object, and wherein a second attribute includes information to resolve validity of ownership of the storage object among the plurality of nodes. The owner node communicates changes to be made to the local versions of the attributes at other nodes of the plurality of nodes. A second node of the plurality of nodes requests ownership of the storage object. The first attribute is updated to designate the second node of the plurality of nodes as the owner node, in response to determining from the second attribute that the validity of ownership of the storage object allows the second node to inherit ownership of the storage object once the first node surrenders ownership of the storage object.

Propagation of updates for attributes of a storage object from an owner node of the storage object to other nodes

Provided are a method, system, and article of manufacture, wherein a plurality of nodes coupled over a network are maintained, wherein a first node of the plurality of nodes initially owns a plurality of storage objects accessible via the network. A second node preemptively takes ownership of a selected storage object of the plurality of storage objects, in response to the first node becoming unavailable, wherein other nodes of the plurality of nodes can request ownership of the selected storage object while the first node is unavailable. A determination is made that the first node that originally owned the selected storage object has become available after a period of unavailability. A determination is made of the current owner node of the selected storage object, wherein the current owner node is one of the second node and the other nodes. The selected storage object is updated at the first node by the current owner node of the selected storage object, prior to allowing the first node to act on or assume ownership of the selected storage object.

Transfer of ownership of a storage object in response to an original owner node becoming available after a period of unavailability

Provided are a method, system, and article of manufacture that maintain a plurality of nodes coupled over a network, wherein a first node of the plurality of nodes initially owns a plurality of storage objects accessible via the network. A serialization lock is maintained in association with the plurality of storage objects, wherein the first node becomes unavailable. A second node of the plurality of nodes determines that the second node is to preemptively assume ownership of a selected storage object of the plurality of storage objects initially owned by the first node that has become unavailable. The second node acquires the serialization lock. The second node secures ownership of the selected storage object, subsequent to acquiring the serialization lock.

Securing transfer of ownership of a storage object from an unavailable owner node to another node

An apparatus, system, and method are disclosed for selecting a source cluster in a distributed storage configuration. A measurement module measures system factors for a plurality of clusters over a plurality of instances. The clusters are in communication over a network and each cluster comprises at least one tape volume cache. A smoothing module applies a smoothing function to the system factors, wherein recent instances have higher weights. A lifespan module calculates a mount-to-dismount lifespan for each cluster from the smoothed system factors. A selection module selects a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

Selecting a source cluster by measuring system factors, calculating a mount-to-dismount lifespan, and selecting the source cluster in response to the lifespan and a user policy

An increase in the efficiency of copy operations in a data storage system is provided. A host device directs that a volume of customer data be stored through one of two virtual storage servers on media in an associated library. An initiating controller queues a request to copy the volume to a second library. If the request is not processed by the initiating controller within a predetermined period, the initiating controller broadcasts the copy request to other controllers, each of which queues the request. After one of the controllers processes the request and the volume has been copied to the second library, the processing controller broadcasts a message to the other controllers which then dequeue the copy request from their respective copy queues. Upon receipt of such message, the initiating controller transmits a message to the host device that the volume was successfully stored and copied.

Self-optimizing workload distribution among virtual storage controllers

An apparatus and method are disclosed to manage one or more reserved volume serial numbers in a virtual library grid. The method supplies a virtual library grid comprising one or more virtual library clusters, wherein each of said one or more virtual library clusters comprises a management interface node, at least one virtualization node, at least one management node, at least one library manager node, at least one direct access storage device, one or more information storage media, and at least one data storage device to read information from and to write information to the one or more information storage media. The method then reserves in each of the one or more virtual library system clusters one or more volume serial numbers, and establishes a time out period. The method then determines if the time out period has expired. If the time out period has expired, the method unreserves in each of the one or more virtual library clusters the one or more volume serial numbers.

Apparatus and method to manage one or more reserved volume serial numbers in a virtual library grid

A dynamic selection of a communications link between a storage controller and one of a plurality of storage servers is provided so as to enhance the performance of the storage system. The storage system includes a plurality of storage devices, a corresponding plurality of virtual storage controllers coupled to the storage devices and a virtual storage controller. The virtual storage controller includes a host interface, a server interface, a stored table and a processing unit. The stored table includes a list identifying the active links between the server interface and each virtual storage server and, for each active link, a corresponding performance value. The processing unit is programmed to receive from a host a request for a storage operation on selected data, determine which active link has the most favorable performance value and direct that the requested storage operation be performed by the virtual storage server coupled to the virtual storage controller over the active link having the most favorable performance value.

Autonomic performance tuning of a distributed virtual storage server

A data storage library system with a plurality of data storage libraries and at least one director accesses a redundant copy of an identifiable data volume, employing the idle time status of each library to balance the workload. The data volumes may be stored in cache storage and backing storage which have different levels of access speed. A director requests a data volume, and the library controller of each library determines its current idle time status, providing the idle time status to the requesting director when it provides a synchronization token directly associated with the requested data volume. The token indicates the update level and the access level of the data volume in the library. The director, upon determining from the tokens that at least two of the copies of the data volume are at the same fastest available access level, compares the provided idle time status of the libraries storing those copies, and accesses the data volume from the library having the greater idle time status.

Balanced access to data volumes with redundant copies stored in data storage libraries

A particularly efficient storage system makes redundant copies of data on tape by waiting until a rewind/unload command is received and then copying stored data objects en masse, thereby benefitting from any data compression used to store the objects along with the lower overhead for the copy operation. This copy operation is more efficient than the data objects' original storage, because there is less overhead in copying the data objects en masse; also, there is less data to copy if the data objects were stored with any data compression. Efficiency may be further increased by dividing the task of originally writing the data objects between paired tape drives, and then exchanging data objects between the tape drives upon issuance of the rewind/unload command.

Backup copy achieving improved efficiency by waiting until rewind/unload command for bulk data copy

Disclosed is a system for managing data in cache. A list of data entries in a first memory area has a first end and a second end, such as a most recently used (MRU) end and least recently used (LRU) end. A first pointer addresses a data entry in the list and a second pointer addresses another data entry in the list that is not at the first and second ends. Data from a second memory area is provided to add to the list. A determination is made as to whether the provided data to add to the list is one of a first type and second type of data, such as sequentially accessed data or non-sequentially accessed data. The provided data is stored in the first memory area as a new data entry in the list. The first pointer is modified to address the new data entry after determining that the provided data is of the first type. After determining that the provided data is of the second type, the second pointer is processed to determine where to add the new data entry to the list between the first and second ends.

Method and system for managing data in cache

Disclosed is a cache management scheme using multiple data structure. A first and second data structures, such as linked lists, indicate data entries in a cache. Each data structure has a most recently used (MRU) entry, a least recently used (LRU) entry, and a time value associated with each data entry indicating a time the data entry was indicated as added to the MRU entry of the data structure. A processing unit receives a new data entry. In response, the processing unit processes the first and second data structures to determine a LRU data entry in each data structure and selects from the determined LRU data entries the LRU data entry that is the least recently used. The processing unit then demotes the selected LRU data entry from the cache and data structure including the selected data entry. The processing unit adds the new data entry to the cache and indicates the new data entry as located at the MRU entry of one of the first and second data structures.

Method and system for managing data in cache using multiple data

A data processing system executes a plurality of processes in parallel. The processes access shared user records stored in memory serially. One or more data structures are used to track access by processes to the data structures and to the user records. Responsive to a fault occurring in a first parallel process, a signal is given the remaining parallel processes indicating the failure. Responsive to the failure, it is determined if the faulting parallel process had access to one of the data structures. Depending upon the circumstances of access, integrity of the user records may be assumed and the record validated. The data structures themselves can be examined to determine if the contents of memory is reliable. Typically the data structures support use of the memory as a cache.

Method of validating memory structures in data processing systems

A storage controller having additional cache memory and a system for recovering from failure and reconfiguring a control unit thereof in response thereto. The inventive controller includes a first cluster for directing data from a host computer to a storage device and a second cluster for directing data from a host computer to a storage device. A first cache memory is connected to the first cluster and a second cache memory is connected to the second cluster. A first nonvolatile memory is connected to the second cluster and a second nonvolatile memory is connected to the first cluster. Data is directed to the first cache and backed up to the first nonvolatile memory. The second cache is similarly backed up by the second nonvolatile memory. In the event of failure of the first cache memory, data is directed to the second cache and backed up in the second nonvolatile memory.

Growing community of inventors

Tucson, AZ, United States of America

Mark Albert Reid