HOST-DEFINED I/O STORAGE CONTROL BASED ON APPLICATION INSTANCE AND/OR CONTROL GROUP

    公开(公告)号:US20200034038A1

    公开(公告)日:2020-01-30

    申请号:US16045889

    申请日:2018-07-26

    Abstract: I/O storage processing behavior is controlled on a storage system based on an application instance of an I/O request. I/O storage processing behavior may be associated with application instances on a host by creating a table or other data structure that assigns identifiers to each application instance and specifies associate behavior(s) with the application instance. The table may associate behavior to application instances based on logical groupings of application instances, which may be based on existing defined groups or other logical entities, for example, a container or VM. A host may communicate the table to a storage system, and the storage system may store information from the table. An I/O request on the host may be tagged with an ID from the table, and sent to the storage system, which controls the I/O storage processing of the I/O request according to the tagged ID.

    HOST DEVICE WITH MULTI-PATH LAYER CONFIGURED FOR DETECTION AND REPORTING OF PATH PERFORMANCE ISSUES

    公开(公告)号:US20200021654A1

    公开(公告)日:2020-01-16

    申请号:US16035785

    申请日:2018-07-16

    Abstract: An apparatus in one embodiment comprises a host device configured to communicate over a network with a storage system comprising a plurality of storage devices. The host device comprises a set of input-output queues and a multi-path input-output driver configured to select input-output operations from the set of input-output queues for delivery to the storage system over the network. The multi-path input-output driver is further configured to send a predetermined command to the storage system over each of a plurality of paths from the host device to the storage system, to monitor a response time for the predetermined command on each of the paths, and to detect a performance issue with at least a given one of the paths based at least in part on the monitored response time. The predetermined command illustratively comprises a Small Computer System Interface (SCSI) “immediate” command of a particular type, such as a Test Unit Ready (TUR) command.

    Host device configured to automatically discover new paths responsive to storage system prompt

    公开(公告)号:US10476960B1

    公开(公告)日:2019-11-12

    申请号:US15967632

    申请日:2018-05-01

    Abstract: An apparatus in one embodiment comprises a host device configured to communicate over a network with a storage system comprising a plurality of storage devices. The host device comprises a set of input-output queues and a multi-path input-output driver configured to select input-output operations from the set of input-output queues for delivery to the storage system over the network. The multi-path input-output driver is further configured to receive per-host path change information from the storage system indicating a change in at least one path from the host device to the storage system, to automatically initiate a path discovery scan in the host device based at least in part on the received per-host path change information, and to add one or more new paths identified in the path discovery scan to a set of active paths for the host device. The per-host path change information received from the storage system may comprise a value of a path change counter maintained by the storage system for the host device.

    Relocation target selection
    44.
    发明授权

    公开(公告)号:US10416929B1

    公开(公告)日:2019-09-17

    申请号:US15662514

    申请日:2017-07-28

    Abstract: Described embodiments provide systems and methods for operating a storage system. A portion of data currently on a source Logical Unit (LUN) of a disk array of a plurality of disk arrays is written to a destination LUN of a disk array of a plurality of disk arrays. The destination LUN of a disk array of the plurality of disk arrays to write the data to is established based on an efficiency indicator associated with the destination LUN of a disk array of the plurality of disk arrays. The writing at least a portion of data currently on a source LUN of a disk array of a plurality of disk arrays may be performed as part of a replication process or as part of a relocation process.

    Managing registration and reservation information in an active-active configuration

    公开(公告)号:US10409838B1

    公开(公告)日:2019-09-10

    申请号:US15074037

    申请日:2016-03-18

    Abstract: Described are techniques for maintaining and using registration and reservation information in an active-active configuration that provides automated data replication of a logical device. Various commands may be issued from one or more hosts to read and modify registration and reservation information in the active-active configuration. Processing is performed to service the various commands and provide a host or client with a consolidated logical view of a data storage system including the logical device where the host is allowed to issue I/Os over multiple paths to the logical device. Depending on the command, such processing may include replicating or mirroring reservation state information between two data storage systems each including a different device configured as the logical device.

    Dynamic front end connectivity optimizations

    公开(公告)号:US10148483B1

    公开(公告)日:2018-12-04

    申请号:US14870718

    申请日:2015-09-30

    Abstract: Described are techniques for determining connectivity. Statistics are received regarding components of the data storage system including any of a target port, a front end adapter, and a device. It is determined in accordance with inputs whether to modify existing connectivity between an initiator set of one or more initiators and a target set of one or more target ports of the data storage system. The inputs include the one or more statistics and one or more adjustment criteria. Responsive to determining to modify the existing connectivity, first processing makes modification(s) to the existing connectivity including any of: modifying an access state associated with a target port over which a device is exposed to an initiator, and adding or removing connectivity between the initiator and another target port of the data storage system where a device is exposed to the initiator over the another target port.

    Method and system for multi-tier all-flash array

    公开(公告)号:US10042751B1

    公开(公告)日:2018-08-07

    申请号:US14871344

    申请日:2015-09-30

    Abstract: Example embodiments of the present invention relate to a method, a system, and a computer program product for tiering metadata. The method includes selecting a consecutive range of addresses of a logical device having a parent data structure associated therewith maintaining a first set of hash values at a first granularity of the logical device. A second hash value then may be calculated over the consecutive range of addresses of the logical device at a second granularity of the logical device and inserted into a child data structure associated with the parent data structure. Entries in the parent data structure at the first granularity for the consecutive range of addresses then may be freed in favor of the second hash value at the second granularity for the consecutive range of addresses inserted into the child data structure, for storing hash values for other addresses of the logical device.

    Efficient data marker representation

    公开(公告)号:US09946485B1

    公开(公告)日:2018-04-17

    申请号:US15386042

    申请日:2016-12-21

    Abstract: A changed block bitmap of a volume of storage is represented by a data marker that includes an offset to a first dirty block and an offset to a last dirty block. The blocks between the first and last dirty blocks are represented by contiguous clean block identifiers, contiguous dirty block identifiers, and contiguous mixed clean and dirty block identifiers. The contiguous clean block identifiers and contiguous dirty block identifiers use one or two bits to indicate identifier type. The remaining bits indicate a numerical count of contiguous clean or dirty blocks. The contiguous mixed clean and dirty block identifiers may use one bit per block to indicate change status. The data marker may be smaller than a corresponding changed block bitmap.

Patent Agency Ranking