公开(公告)号：GB2584064B

公开(公告)日：2021-04-14

申请号：GB202014835

申请日：2019-02-15

Applicant: IBM

Inventor： MATTHEW BORLICK ,  LOKESH GUPTA ,  KEVIN JOHN ASH ,  KARL ALLEN NIELSEN

IPC: G06F11/10 ,  G06F3/06

Abstract: In one embodiment, virtual storage drives are allocated to RAID arrays so that no two virtual storage drives of a RAID array are mapped to the same physical storage drive. In another aspect, error handling routines are limited to virtual storage drives impacted by an error in a physical storage drive so that virtual storage drives of the physical storage drive not impacted by the error are bypassed. In yet another aspect, cache operations to a target virtual storage drive may be throttled as a function of both a limit imposed on cache operations directed to the RAID array to which the virtual storage drive is allocated, and a separate limit on cache operations directed to a group of virtual storage drives which are mapped to the same physical storage drive as the target virtual storage drive. Other features and aspects may be realized, depending upon the particular application.

公开(公告)号：GB2580263A

公开(公告)日：2020-07-15

申请号：GB202004106

申请日：2018-08-20

Applicant: IBM

Inventor： LOKESH M GUPTA ,  MATTHEW BORLICK

IPC: G06F12/08

Abstract: An example method for dynamic data relocation using cloud based ranks comprises monitoring accesses to data stored on a plurality of local ranks of an enterprise storage system; identifying data which has not been accessed for a predetermined amount of time based on the monitored accesses; and moving the data which has not been accessed for the predetermined amount of time to one or more cloud based ranks of the enterprise storage system, wherein each cloud based rank comprises storage space on one or more cloud storage devices, the storage space on the one or more cloud storage devices mapped to corresponding virtual local addresses that are grouped as a virtual local rank.

公开(公告)号：GB2580259A

公开(公告)日：2020-07-15

申请号：GB202004071

申请日：2018-08-20

Applicant: IBM

Inventor： LOKESH M GUPTA ,  MATTHEW BORLICK

IPC: G06F12/08

Abstract: A computer-implemented method for thin provisioning using cloud based ranks comprises determining a total amount of unused physical storage space for all of a plurality of local ranks associated with a storage controller; comparing the total amount of unused physical storage space to a first threshold; in response to determining that the total amount of unused physical storage space is less than the first threshold, creating one or more cloud based ranks. Creating each of the one or more cloud based ranks comprises allocating storage space on one or more corresponding cloud storage devices via a cloud interface; mapping the allocated storage space to corresponding virtual local addresses; and grouping the virtual local addresses as a virtual local rank associated with the storage controller.

公开(公告)号：GB2579992A

公开(公告)日：2020-07-08

申请号：GB202003940

申请日：2018-09-06

Applicant: IBM

Inventor： LOKESH MOHAN GUPTA ,  MATTHEW BORLICK

IPC: G06F12/08

Abstract: A computer-implemented method for utilizing cloud storage as a rank comprises allocating storage space on one or more cloud storage devices via a cloud interface; mapping the allocated storage space to corresponding virtual local addresses; grouping the virtual local addresses to create one or more virtual local ranks from the allocated storage space on the one or more cloud storage devices; converting local data access requests for the one or more virtual local ranks to cloud data access requests configured for the cloud interface; and communicating the cloud data access requests to the one or more cloud storage devices via the cloud interface.

公开(公告)号：GB2599322B

公开(公告)日：2023-02-08

申请号：GB202200083

申请日：2020-06-11

Applicant: IBM

Inventor： LOKESH MOHAN GUPTA ,  MATTHEW BORLICK ,  KARL ALLEN NIELSEN ,  MICAH ROBISON

IPC: G06F3/06

Abstract: A method for dynamically altering the performance class of multiple storage drives is disclosed. In one embodiment, such a method monitors, within a storage environment, characteristics (e.g., age, wear, etc.) of multiple storage drives. Each storage drive has a performance class associated therewith. Based on the characteristics, the method periodically modifies the performance class of the storage drives. The method then reorganizes the storage drives within various storage groups (e.g., RAID arrays, storage tiers, workloads, etc.) based on their performance class. For example, the method may place, as much as possible, storage drives of the same performance class within the same storage groups. A corresponding system and computer program product are also disclosed.

公开(公告)号：GB2586398B

公开(公告)日：2022-06-08

申请号：GB202015329

申请日：2019-02-27

Applicant: IBM

Inventor： LOKESH MOHAN GUPTA ,  MATTHEW BORLICK ,  ROGER HATHORN

IPC: G06F17/00

Abstract: In one aspect of the present description, safe data commit scan operations of individual data storage systems of a distributed data storage system may be synchronized to reduce the occurrence of reductions in input/output (I/O) response times. In one embodiment, a set of safe data commit scan operations of the individual data storage systems of a distributed data storage system are synchronously timed to substantially overlap in time within a single synchronized safe data commit scan set interval to reduce or eliminate the occurrences of reductions in input/output response times outside the synchronized safe data commit scan set interval. Other features and aspects may be realized, depending upon the particular application.

公开(公告)号：GB2599322A

公开(公告)日：2022-03-30

申请号：GB202200083

申请日：2020-06-11

Applicant: IBM

Inventor： LOKESH MOHAN GUPTA ,  MATTHEW BORLICK ,  KARL ALLEN NIELSEN ,  MICAH ROBISON

IPC: G06F12/00

Abstract: A method for dynamically altering the performance class of multiple storage drives is disclosed. In one embodiment, such a method monitors, within a storage environment, characteristics (e.g., age, wear, etc.) of multiple storage drives. Each storage drive has a performance class associated therewith. Based on the characteristics, the method periodically modifies the performance class of the storage drives. The method then reorganizes the storage drives within various storage groups (e.g., RAID arrays, storage tiers, workloads, etc.) based on their performance class. For example, the method may place, as much as possible, storage drives of the same performance class within the same storage groups. A corresponding system and computer program product are also disclosed.

公开(公告)号：GB2591424A

公开(公告)日：2021-07-28

申请号：GB202106810

申请日：2019-09-30

Applicant: IBM

Inventor： LOKESH GUPTA ,  KYLER ANDERSON ,  KEVIN JOHN ASH ,  MATTHEW BORLICK

IPC: G06F12/08

Abstract: A machine learning module receives inputs comprising attributes of a storage controller, where the attributes affect performance parameters for performing stages and destages in the storage controller. In response to an event, the machine learning module generates, via forward propagation, an output value that indicates whether to fill holes in a track of a cache by staging data to the cache prior to destage of the track. A margin of error is calculated based on comparing the generated output value to an expected output value, where the expected output value is generated from an indication of whether it is correct to fill holes in a track of the cache by staging data to the cache prior to destage of the track. An adjustment is made of weights of links that interconnect nodes of the plurality of layers via back propagation to reduce the margin of error.

公开(公告)号：GB2585320B

公开(公告)日：2021-06-09

申请号：GB202017405

申请日：2019-04-01

Applicant: IBM

Inventor： LOKESH GUPTA ,  MATTHEW BORLICK ,  KEVIN JOHN ASH

IPC: G06F12/0866 ,  G06F3/06 ,  G06F12/0895

Abstract: Provided are a computer program product, system, and method for using track locks and stride group locks to manage cache operations. A group of tracks from the storage devices are stored in a cache. Exclusive track locks for tracks in the group in the cache are granted for writes to the tracks in the group in the cache, wherein exclusive track locks can be simultaneously held for writes to different tracks in the cache. An exclusive group lock for the group of tracks in the cache is granted to destage the tracks in the group from the cache to the storage devices. The exclusive group lock is released in response to completing the destage of the tracks in the group in the cache to the storage devices.

公开(公告)号：GB2587314A

公开(公告)日：2021-03-24

申请号：GB202100441

申请日：2019-06-20

Applicant: IBM

Inventor： MATTHEW BORLICK ,  KARL ALLEN NIELSEN ,  CLINT HARDY ,  LOKESH GUPTA

IPC: G06F11/07 ,  G06F11/10 ,  G06N3/08 ,  G06N5/00 ,  G06N5/02 ,  G06N20/10

Abstract: Provided are a computer program product, system, and method for using a machine learning module to determine when to perform error checking of a storage unit. Input on attributes of at least one storage device comprising the storage unit are provided to a machine learning module to produce an output value. An error check frequency is determined from the output value. A determination is made as to whether the error check frequency indicates to perform an error checking operation with respect to the storage unit. The error checking operation is performed in response to determining that the error checking frequency indicates to perform the error checking operation.