Hai-Fang Yun

6336164, Jan 1, 2002

May 12, 1999

09/311205

Steven Gerdt - San Jose CA
Juan Li - San Jose CA
Dung Lim Nguyen - San Jose CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1214

711113, 711155, 707206

Aspects for preventing deadlock in a log structured array are described. In an exemplary method aspect, and system for providing same, the method includes monitoring available segments to determine when to reserve an open segment, identifying a process type requesting access to the open segment, and limiting access to one process type for the reserved segment. Limiting access further includes limiting access to a garbage collection process.

6463503, Oct 8, 2002

May 12, 1999

09/311209

Michael Reese Jones - Longmont CO
Juan Li - San Jose CA
Dung Kim Nguyen - San Jose CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1214

711114, 711111, 711112, 711113, 711155, 711157, 711158, 707203, 707204, 707206

Aspects for increasing concurrency during staging and destaging of a log structured array (LSA) are described. In an exemplary method aspect, the method includes determining a process type making a request for access to a locking mechanism associated with an LSA directory entry for a logical track, and determining a lock status for the locking mechanism. Access to the locking mechanism is then provided according to the determined process type and lock status. Further, a destaging process is allowed to obtain an available locking mechanism in a read-write mode, a staging process is allowed to obtain an available locking mechanism in a read-only mode, and a garbage collection process is allowed to obtain an available locking mechanism initially in a read-only mode and subsequently in a read-write mode, wherein increased concurrency results during the staging and destaging processes.

6473830, Oct 29, 2002

Apr 30, 2001

09/846148

Juan Li - San Jose CA
Dung Kim Nguyen - San Jose CA
Aare Onton - Saratoga CA
Kevin Frank Smith - Morgan Hill CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1200

711112, 711114, 711154, 711170

In a storage system comprising an array of storage devices, including a processor and memory, a plurality of logical tracks are organized on the storage devices in segments comprising columns striped across the storage devices. A system and method are provided for storing logical tracks in the storage devices. Sequentially logically related logical tracks are stored together in neighborhoods. Logical tracks of the same neighborhood destaged at the same time are assigned to a single open segment with other logical tracks of the same neighborhood type. The time at which open segments are designated as closed segments to be written to the storage devices is based on performance, disk utilization and memory utilization criteria. Logical tracks are never split between segment columns. Also, attempts are made not to split a neighborhood of logical tracks being destaged together between segment columns.

6490664, Dec 3, 2002

Jun 14, 1999

09/333143

Michael Reese Jones - Longmont CO
Juan Li - San Jose CA
Dung Kim Nguyen - San Jose CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1208

711154, 710 32, 714 24, 707200, 711114

Aspects for handling destage requests during shutdown in a log-structured array storage subsystem are described. In a method aspect, the method includes receiving a shut-down command, and utilizing at least three data structures for tracking destage requests when the shut-down command is received, wherein closing of open segments before completion of the shut-down is ensured. A further method aspect includes maintaining an outstanding requests list and destage requests list, forming a missing requests list based on the contents of the outstanding requests list and destage requests list when a shut-down command occurs, and tracking destage request processing with the outstanding requests list, destage requests list and missing requests list until all destage requests have been successfully completed.

6842836, Jan 11, 2005

Nov 7, 2001

10/005575

Hai-Fang Yun - San Jose CA, US
Leonard McCrigler - San Jose CA, US

Hewlett-Packard Development Company, L.P. - Houston TX

G06F 1200

711167, 711 3, 711151, 711168

A method of caching data. In one embodiment, the method is comprised of filling a cache with incoming data to a first level. The filling is at a rate relative to said incoming data. The method is further comprised of increasing the cache from the first level to an optimum level. Outputting of the incoming data is enabled subsequent to the cache attaining the first level. The method is further comprised of adjusting the level of said cache level concurrent with incoming data and data outputting. This adjusting prevents the level of the cache from exceeding a maximum cache level and prevents the level of the cache from decreasing below the first level, such that smooth and continuously-streaming outputting of said data is provided.

62567055, Jul 3, 2001

Jun 5, 1998

9/092433

Juan Li - San Jose CA
Dung Kim Nguyen - San Jose CA
Aare Onton - Saratoga CA
Kevin Frank Smith - Morgan Hill CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1200
G06F 1300

711112

In a storage system comprising an array of storage devices, including a processor and memory, a plurality of logical tracks are organized on the storage devices in segments comprising columns striped across the storage devices. A system and method are provided for storing logical tracks in the storage devices. Sequentially logically related logical tracks are stored together in neighborhoods. Logical tracks of the same neighborhood A0 destaged at the same time are assigned to a single open segment with other logical tracks of the same neighborhood type. The time at which open segments are designated as closed segments to be written to the storage devices is based on performance, disk utilization and memory utilization criteria. Logical tracks are never split between segment columns. Also, attempts are made not to split a neighborhood of logical tracks being destaged together between segment columns.

61516856, Nov 21, 2000

May 15, 1998

9/079868

Juan Li - San Jose CA
Dung K. Nguyen - San Jose CA
Mien Shih - Saratoga CA
Hai-Fang Yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1100

714 6

In a storage system comprising an array of storage devices including a processor and memory, a plurality of data blocks are organized on the storage devices in segments striped across the storage devices. A main directory, stored in memory, contains the location on the storage device of each of the data blocks. For each segment a segment directory having information on the data blocks in the segment is stored on the storage devices. When a damaged segment directory is detected, a checkpoint of the main directory is written to the storage devices. A list is maintained of damaged segment directories such that segments on the damaged segment directories list are not garbage collected. Following a main directory recovery procedure, the damaged segment directories are reconstructed using the main directory.

60527994, Apr 18, 2000

May 15, 1998

9/079913

Juan Li - San Jose CA
Dung K. Nguyen - San Jose CA
Mien Shih - Saratoga CA
Hai-Fang yun - San Jose CA

International Business Machines Corporation - Armonk NY

G06F 1100
G06F 1200

714 13

In a storage system comprising an array of storage devices including a processor and memory, a plurality of data blocks are organized on the storage devices in segments striped across the storage devices. A main directory stored in memory contains the location on the storage device of each of the current data blocks. A method is provided for recovering a main directory in the event that the main directory becomes damaged. A segment directory having information on the data blocks in the segment is stored as part of the segment on the storage devices. Periodically a checkpoint of the main directory is written to the storage devices. A list is maintained of closed segments written to the storage devices since the checkpoint directory was written to the storage devices. During a main directory recovery procedure, the checkpoint of the main directory is read into memory, and for each segment that is indicated by the closed segments list as having been closed since the checkpoint of the main directory was written to the disk, the information in the corresponding segment directory is used to update the checkpoint directory.