International Business Machines Corporation

Ronald Maynard Kern

Gregory E McBride

Robert Nelson Crockett

David Michael Shackelford

William Frank Micka

Robert Frederic Kern

Mark Anthony Sovik

John Edward Kulakowski

Thomas Charles Jarvis

Robert Wagner

Mark Alan Haye

Claus William Mikkelsen

Donald Reynold Blea

William J Flannagan

Harry M Yudenfriend

Jeffrey Mark Nick

John Glenn Thompson

David Randall Blea

David B Petersen

Robert Wesley Shomler

James Mitchell Ratliff

Douglas Michael Zobre

Stephen C West

Robert Vaughn Geiner

Larry Richard Perry

Lawrence E Eastridge

Harold Glen Slone

Debra L Jaworski

Method, apparatus and program product are provided for rebuilding faulty metadata in a storage controller coupled to a host device. Faulty metadata may include metadata which no longer matches the associated customer data tracks stored on a DASD or other storage device. When an error in metadata is describing a first customer track is detected, a range of other customer tracks in which the error is also likely to have occurred is identified. The metadata tracks associated with the first customer track and the other customer tracks are selected to be invalidated. A command is received through a host interface, and the specified metadata tracks are invalidated while the controller remains on-line with the host and continues to process other host I/O operations. Subsequently, the invalidated metadata tracks are rebuilt. The disclosed method, apparatus and program product invalidate the faulty metadata with reduced impact on normal host/controller I/O operations.

Invalidation of storage control unit cache metadata

Method, apparatus and program product are provided for the invalidation of faulty metadata in a storage controller coupled to a host device. Faulty metadata may include metadata which no longer matches the associated customer data tracks stored on a DASD or other storage device. When faulty metadata is detected, metadata tracks are selected to be invalidated. A command is received through a host interface, without the controller being taken off-line from the host, and the specified metadata tracks are invalidated. Subsequently, the invalidated metadata tracks are rebuilt. The disclosed method, apparatus and program product invalidate the faulty metadata with reduced impact on normal host/controller I/O operations.

In a system including a host, a primary storage subsystem coupled to the host, a cache coupled to the host and separate from the primary storage system, a secondary storage subsystem, and a data mover coupling the primary and secondary storage systems, data is temporarily cached for future storage in the primary storage subsystem so as to preserve timestamp information and maintain data consistency for asynchronously mirroring the data at a secondary subsystem.

Adjusting timestamps to preserve update timing information for cached data objects

Backup storage is resynchronized to primary storage, ensuring that any new updates received during resynchronization are applied in the proper order relative to resynchronization data. Under normal operations, a data mover mirrors data stored in primary storage to backup storage. If an error condition arises, preventing mirroring, the data mover stores newly received data in primary storage without mirroring the data to backup storage. The data mover also identifies this data in an update map. When the error condition ends, the data mover performs a static resynchronization process, serving to update the backup storage with the un-mirrored data, identified in the update map. When new data is received during static resynchronization, a dynamic resynchronization process is invoked to accurately process the updates. Dynamic resynchronization ensures that newly received data records are copied to backup storage in the proper order (if at all) with respect to versions of the same data being processed by static resynchronization.

System and method for dynamically resynchronizing backup data

A multi-session data storage facility coordinates similar status changes across all sessions at appropriate times. In each session, a data mover implements data mirroring by copying updates from the primary storage to the secondary storage. A master data set, accessible by all sessions, includes a common area for posting commands and/or errors to all sessions and individual session areas for posting errors occurring in particular sessions. When a data mover receives a host initiated command, this data mover becomes a &ldquo;master.&rdquo; Initially, the master data mover lists the command in the master data set's common area. Whenever other, &ldquo;slave&rdquo; data movers notice a command in the master data set's common area, they respond by suspending their formation of consistency groups and responding with a &ldquo;ready&rdquo; message. The &ldquo;ready&rdquo; message comprises the slave data mover's consistency form time (i.e., the earliest of most-recent update time from all controllers in that session). Accordingly, after posting the command, the master data mover reviews the slave data movers' responses. If these responses indicate all sessions can execute the command at a requested consistent time, then the master data mover instructs the slave data movers to execute the command by entering an execute instruction in the master data set's common area. Otherwise, the master data mover enters a cancel operation in the master data set's common area.

Coordinating remote copy status changes across multiple logical sessions to maintain consistency

A system, method, and program for maintaining data consistency among updates to data storage areas are provided. Each update has an update time the update was made. There are multiple groups of data storage areas. For each group, updates to the data storage area in the group are stored in a journal for storing updates to the group, wherein there are multiple journals. An indication is made in a memory area for each group of a group update time comprising a most recent update time of the updates in the group. The update time for each update in the group is not greater than the group update time. A determination is made of a minimum group update time across all the groups. At least one update is applied to storage if the update time for the update does not exceed the minimum group update time. The data storage areas may be partitioned among a plurality of computer systems, and may be maintained in at least one primary and secondary storage. The data storage areas may also comprise volumes, with each group including at least one volume. The groups of updates may also be maintained in a journal.

Method, system and program for maintaining data consistency among updates across groups of storage areas using update times

A storage-and-host-controller-managed outboard data management tool is described wherein the host controller defines logical paths between various storage controllers and data storage devices, and the storage controller manages the movement of data to and from the various data storage devices while only sending data to the host processorg if necessary.

Host storage management control of outboard data movement

Disclosed is a system for maintaining consistency of data across storage devices. A cut-off time value is provided to the system. The system then obtains information on data writes to a first storage device, including information on time stamp values associated with the data writes indicating an order of the data writes to the first storage device. At least one group of data writes having time stamp values earlier in time than the cut-off time value is then formed. The system then transfers the data writes in the groups to a second storage device for storage therein.

Multi-environment data consistency

Disclosed is a system for managing pages in a volatile memory device for data transfer operations between a first storage area and a second storage area. The first storage area is queried to determine a number of data sets to include in a data transfer operation. A number of pages in the volatile memory device needed for the data transfer operation is then determined. A determination is then made as to whether the number of pages needed for the data transfer operation is greater than available fixed pages in a pool of pages. Available fixed pages in the pool are allocated to the data transfer operation after determining that the number of pages needed to process the data transfer operation is less than or equal to the available fixed pages in the pool. Alternatively, after determining that the number of pages needed to process the data transfer operation is greater than the available fixed pages in the pool, all the fixed pages in the pool and additional fixed pages are allocated to the data transfer operation.

Management of fixed pages in memory for input/output operations

A storage-and-host-controller-managed outboard data management tool is described wherein the host controller defines logical paths between various storage controllers and data storage devices, and the storage controller manages the movement of data to and from the various data storage devices while only sending data to the host processor if necessary.

Host storage management control of outboard data movement using push-pull operations

A method, apparatus, and article of manufacture for performing a point-in-time backup using a snapshot function. Work space on a work virtual volume is obtained to hold the snapshot copy of the source data being copied from a source virtual volume. Updates to the source virtual volume are suspended during the snapshot function, so as to provide a logically consistent view of the source virtual volume at a common point in time. Pointers are copied from a virtual track table of the source virtual volume to a virtual track table of the work virtual volume without actually moving any data on data storage devices referenced by those pointers. As a result, the snapshot function provides a method for copying the source virtual volume to the work virtual volume very quickly. Upon completion of the snapshot function, updates to the source virtual volume may be resumed. Subsequently, a backup is performed in the usual manner, except that the backup retrieves the source data from the work virtual volume rather than the source virtual volume.

Method for implementing point-in-time copy using a snapshot function

A data storage system includes an outboard data transfer module that directs data transfer operations without requiring substantial host involvement. This subsystem includes a host computer, one or more digital data storage devices, one or more device controllers interposed between the host computer and the storage devices, and a outboard data manager coupled to the host and the device controllers. Data transfer operations are initiated when the outboard data manager receives a data transfer request from the host. The request identifies the type of operation, resulting target format, source data residing on one or more source devices, and one or more target devices. In response to this request, the outboard data manager retrieves a copy of the source data from the source device(s), formats target data according to host instruction, and then transmits the copy to the target device(s) for storage therein. Thus, source data is copied from source to target storage device(s) via the outboard data manager without burdening the host by routing data through it. Some exemplary data transfer operations include copy, migrate, recall, backup, restore, move, defragment, reorganize, and the like.

Data storage system with outboard physical data transfer operation utilizing data path distinct from host

A primary data mover monitors both consistency time and idle time in a system that performs continuous, asynchronous, extended remote copying between primary and remote processors, and manages both with accuracy and consistency. The primary data mover detects system activity levels and manages data accuracy for the extended remote copying in both active and idle systems.

Extended remote copying system for reporting both active and idle

A method, apparatus, and article of manufacture for dynamic change management in an extended remote copy system. A primary data mover monitors changes made to a primary data volume resulting from write I/O operations, wherein the changes are reflected in a data structure maintained by the primary data mover. The primary data mover performs a multi-level analysis on the data structure to discard a portion of the changes therein, so that the changes that have been applied to a secondary data volume are discarded from the data structure and the changes that have not been applied to the secondary data volume are maintained in the data structure.

Growing community of inventors

Tucson, AZ, United States of America

Ronald Maynard Kern