International Business Machines Corporation

Richard K Treiber

Spencer W Ng

Ronald L Obermarck

Mario Yamaguchi

Jonathan Mark Haswell

William Frank Micka

Warren Keith Stanley

Chandrasekaran Mohan

Jaishankar Moothedath Menon

Robert Frederic Kern

John Davis Palmer

Robert Wesley Shomler

Claus William Mikkelsen

Margaret Ann Dong

Kurt A Shoens

A storage-controller-managed outboard incremental backup/restore of data. An indicator is assigned to an associated portion of data stored in a data storage device, wherein the indicator resides in a memory of a storage controller coupled to the data storage device. A value of the indicator residing within the memory of the storage controller is set when the associated portion of the data has changed. The associated portion of the data stored in the data storage device is backed using the storage controller when the value of the indicator residing within the memory of the storage controller is set. The value of the indicator residing within the memory of the storage controller is cleared when the associated portion of the data has been backed up.

Storage-controller-managed outboard incremental backup/restore of data

A system manages data transfer between a computer, a data storage disk, and a data buffer which is associated with the disk. Upon receipt of a write request during data read-ahead from the disk to the buffer, the system preempts the read-ahead to write the data to the disk only when the target write track on the disk is different from the track being read ahead into the buffer. Otherwise, the system permits read-ahead to continue until the target write sector is positioned beneath the disk drive head. Also, upon receipt of a read request during cache writing from the buffer to the disk, the system immediately satisfies the read request prior to completing the cache writing when the requested read data is contained in the buffer. Moreover, the system provides for dynamic switching between a data read-ahead preempt policy and a data read-ahead no-preempt policy based upon parallel monitoring and comparison of data request service times under both policies.

Method and apparatus for computer disk drive buffer management

A disaster recovery system providing asynchronous remote data shadowing between a primary and a secondary site uses a pair of track arrays for each primary data storage device of a remote copy pair to monitor which record updates are in transit between the primary site and the secondary site. A host processor at the primary site of the disaster recovery system transfers a sequentially consistent order of copies of record updates to the secondary site for back-up purposes. The copied record updates are stored on the secondary data storage devices which form remote copy pairs with the primary data storage devices at the primary site. One track array, the active track array, is used to set elements according to which tracks on the primary data storage device receive record updates from the host processor at the primary site. The other track array, the recovery track array, designates which record updates comprise the copied record updates currently transferred from the primary site to the secondary site for data shadowing and is used for recovery should an error interrupt the transfer. The track arrays are toggled once the consistency group transfer completes, the recovery track array becoming the active track array and the active track array becoming the recovery track array.

Asynchronously shadowing record updates in a remote copy session using

Method and apparatus for computer disk drive management

Change processing of a replica database is accomplished by separating redo records obtained from the transaction log of a primary database into respective queues. The redo records are separated such that all transaction records for a unit of transfer (page) of the primary database are placed on the same queue in log sequence. Each queue is linked exclusively to one of a plurality of parallel queue servers. Each queue server applies to the replica database the redo records in the queues which it exclusively serves. The replica database is thereby made consistent with the primary data by a lock-free updating mechanism which services the pages of the replica database in parallel.

Concurrently applying redo records to backup database in a log sequence

A method for expanding and contracting a resource pool dynamically, i.e. in response to demand for the resources, is disclosed. Resources are allocated for use from a single master pool 22. The overall pool of available resources is expanded by creating an extension pool and adding its resources to the master pool. The master pool can theoretically be expanded to the limits of the number of resources available in the system. To contract the pool dynamically, first one of the extension pools is selected to be freed (FREE.sub.-- POOL 66). Those resources owned by the freed pool but still in the master pool (i.e., not in use) are moved to the extension pool, preventing their reuse. Resources in use which are owned by the freed pool are labeled by changing their home address 52 so that they will be returned to the extension pool instead of to the master pool. Once all such resources have been returned to the freed extension pool it is deleted, thereby contracting the overall pool of resources. Pseudocode implementations of the procedures for manipulating storage buffers and buffer pools are included.

Dynamic resource pool expansion and contraction in multiprocessing

A method for increasing throughput of N-way central electronic complexes concurrently executing processes to selectively lockable data resources while maintaining coherency among replicates of the information state of any accessed resource. Throughput is increased by overlapping a resource lock request with the processing incidental to accessing the resource. Processes are granted locks to portions of the resource based on the interest state held by the central electronic complex in the resource. Messages are attached to locked resources and are distributed to the appropriate complexes by virtue of a list maintained by the central electronic complex holding sole interest in the resource. The central electronic complex distributing the messages must receive acknowledgement of the message reception before releasing the resource.

Method for lock management, page coherency, and asynchronous writing of

A method for serializing process access to shared resources utilizing low-level atomic functions to maintain control structures in which noncontiguous words must be modified at two different times. The atomic functions require an initiation operation partitionable from a completion operation. Each process requesting access to the shared resource performs the initiation operation, which begins modification of a control structure. The completion operation may be performed by the original process if there is no resource conflict. If, however, another process is currently modifying the control structure, then the task of performing the completion operation is passed to one of the concurrently-accessing processes.

Growing community of inventors

San Jose, CA, United States of America

Richard K Treiber