International Business Machines Corporation

Kenneth Wayne Todd

Michael T Benhase

Lokesh M Gupta

Kevin J Ash

Brent Cameron Beardsley

Robert Louis Morton

Carol Santich Mellgren

Alfred Emilio Sanchez

Beth Ann Peterson

Thomas Charles Jarvis

Douglas A Martin

Susan Kay Candelaria

David Frank Mannenbach

Steven Robert Lowe

Kurt Allen Lovrien

Olympia Gluck

Matthew J Kalos

Karl Allen Nielsen

Theresa Mary Brown

Gail Andrea Spear

William Frank Micka

Brian Anthony Rinaldi

Warren Keith Stanley

Joseph S Hyde, Ii

Gary William Batchelor

Suguang Li

David Blair Whitworth

Lawrence Carter Blount

Mark L Lipets

Vernon John Legvold

Anthony DeGroff Drumm

Ellen Joanne Grusy

Randall Clay Itskin

Colleen Renee Stouffer

William Dennis Williams

Ronald James Chapman

James D Marwood, Jr

Boyan Zhao

Carol Santich Melgren

Kevin John Ash

Lokesh Mohan Gupta

For data management by a processor device in a computing storage environment, a threshold for an amount of Non Volatile Storage (NVS) space to be consumed by any particular logically contiguous storage space in the computing storage environment is established based on at least one of a Redundant Array of Independent Disks (RAID) type, a number of point-in-time copy source data segments in the logically contiguous storage space, and a storage classification. Establishing the threshold for the amount of NVS to be consumed based on the number of point-in-time copy source data segments in the logically contiguous storage space further includes considering resources needed to perform a Copy Source To Target (CST) operation required prior to the point-in-time copy source data segments being destaged.

NVS thresholding for efficient data management

For data management by a processor device in a computing storage environment, a threshold for an amount of Non Volatile Storage (NVS) space to be consumed by any particular logically contiguous storage space in the computing storage environment is established based on at least one of a Redundant Array of Independent Disks (RAID) type, a number of point-in-time copy source data segments in the logically contiguous storage space, and a storage classification.

A storage controller receives a request to avoid extent conflict checking from a host application executing in a host that is coupled to the storage controller. In response to receiving a request, a host I/O process that executes in the storage controller to satisfy I/O requests from the host application opportunistically releases a track of an extent to a copy services process at an earliest point in time when the track of the extent is longer needed by the host I/O process and is needed by the copy services process.

Management of extent checking in a storage controller during copy services operations

Data operations, requiring a lock, are batched into a set of operations to be performed on a per-core basis. A global lock for the set of operations is periodically acquired, the set of operations is performed, and the global lock is freed so as to avoid excessive duty cycling of lock and unlock operations in the computing storage environment.

Deferred re-MRU operations to reduce lock contention

A system for data management in a computing storage environment includes a processor device, operable in the computing storage environment, that divides a plurality of counters tracking write and discard storage operations through Non Volatile Storage (NVS) space into first, accurate, and second, fuzzy, groups where the first, accurate, group is one of updated on a per operation basis, while the second, fuzzy, group is one of updated on a more infrequent basis as compared to the first, accurate group.

Fuzzy counters for NVS to reduce lock contention

A method for data management in a computing storage environment includes a processor device, operable in the computing storage environment, that divides a plurality of counters tracking write and discard storage operations through Non Volatile Storage (NVS) space into first, accurate, and second, fuzzy, groups where the first, accurate, group is one of updated on a per operation basis, while the second, fuzzy, group is one of updated on a more infrequent basis as compared to the first, accurate group.

A processor, operable in a computing storage environment, for each rank in a storage management device in the computing storage environment, allocates a lower maximum count defined by a predetermined lower maximum count of Task Control Blocks (TCBs) of a rank for performing destage operations, and a higher maximum count of TCBs to be implemented for performing a storage operation, and performs the storage operation using up to the lower maximum count of TCBs, yet only allows those TCBs above the lower maximum count to be allocated for performing the storage operation satisfying at least one criterion.

Management of destage tasks with large number of ranks

Mechanisms for concurrent update of data in cache with destage of data from the cache to disk by a processor device. A second copy of the data is established in the cache and on a cache directory. A first copy of the data and the second copy of the data are adjacently ordered in the cache directory. One of the first and second copies is held for an update operation so as to include a latest data modification, while the remaining copy concurrently is used for a destage operation to disk.

Concurrent update of data in cache with destage to disk

Various embodiments for initializing metadata in a computing storage environment by a processor. A Release Generation Number (RGN) is associated with a volume, and an RGN is associated with a metadata track. Upon a release of storage space in the volume, the RGN associated with the volume is incremented. Upon an initialization of the metadata track, the RGN associated with the metadata track is updated to be consistent in generation with the RGN associated with the volume. Upon an access of the metadata track, the RGN of the metadata track is compared against the RGN of the volume, and the metadata track is initialized if a match is not found.

Deferred metadata initialization

A method, system, and computer program product for grouping and dispatching scans in a cache directory of a processing environment is provided. A plurality of scan tasks is aggregated from a scan wait queue into a scan task queue. The plurality of scan tasks is determined by selecting one of (1) each of the plurality of scan tasks on the scan wait queue, (2) a predetermined number of the plurality of scan tasks on the scan wait queue, and (3) a set of scan tasks of a similar type on the scan wait queue. A first scan task from the plurality of scan tasks is selected from the scan task queue. The scan task is performed.

Grouping and dispatching scans in cache

For cache/data management in a computing storage environment, incoming data segments into a Non Volatile Storage (NVS) device of the computing storage environment are validated against a bitmap to determine if the incoming data segments are currently in use. Those of the incoming data segments determined to be currently in use are designated to the computing storage environment to protect data integrity.

Recovery from cache and NVS out of sync

A processor, operable in a computing storage environment, for each rank in a storage management device in the computing storage environment, allocates a lower maximum count, and a higher maximum count, of Task Control Blocks (TCBs) to be implemented for performing a storage operation, and performs the storage operation using up to the lower maximum count of TCBs, yet only allows those TCBs above the lower maximum count to be allocated for performing the storage operation satisfying at least one criterion.

Growing community of inventors

Tucson, AZ, United States of America

Kenneth Wayne Todd