Write performance distribution monitoring for write operation adaptation

    公开(公告)号:US11422711B1

    公开(公告)日:2022-08-23

    申请号:US17107135

    申请日:2020-11-30

    Abstract: A computing device includes an interface configured to interface and communicate with a dispersed storage network (DSN), a memory that stores operational instructions, and a processing module operably coupled to the interface and memory such that the processing module, when operable within the computing device based on the operational instructions, is configured to perform various operations. For example, the computing device monitors storage unit (SU)-based write transfer rates and SU-based write failure rates associated with each of the SUs for a write request of encoded data slices (EDSs) to the SUs within the DSN. The computing device generates and maintains a SU write performance distribution based on monitoring of the SU-based write transfer rates and the SU-based write failure rates and adaptively adjusts a trimmed write threshold number of EDSs and/or a target width of EDSs for write requests of sets of EDSs to the SUs within the DSN.

    Write response thresholds
    122.
    发明授权

    公开(公告)号:US11418591B1

    公开(公告)日:2022-08-16

    申请号:US16806123

    申请日:2020-03-02

    Abstract: Systems and methods for storing encoded data slices in a storage network involve generating a set of write slice requests relating to the encoded data slices, outputting the set of write slice requests to distributed storage units, receiving write slice responses and determining whether to retry storing the set of encoded data slices based on one or more of the write slice responses. Determining whether to retry storing the set of encoded data slices may involve determining whether less than a write threshold number of favorable responses were received, for example in a time period, and/or whether an unfavorable response was received. Determining whether to retry may also be based on receiving a slow or non-response. For purposes of retrying, the systems and methods may also determine whether to modify the write slice requests, for example by modifying a revision number.

    Optimized Selection of Participants in Distributed Data Rebuild/Verification

    公开(公告)号:US20220214944A1

    公开(公告)日:2022-07-07

    申请号:US17655841

    申请日:2022-03-22

    Abstract: A storage network method begins by receiving a data retrieval request, wherein a data object is segmented into a plurality of data segments, determining a storage configuration associated with the data retrieval request based on an encoded data slice identifier included in the retrieval request and, based on the storage configuration, determining a plurality of storage resources associated with the encoded data slice identifier The mehod continues by determining an estimated performance level for each storage resource of a plurality of storage resources associated with the encoded data slice identifier, determining an expected retrieval performance for the retrieval request and selecting a storage resource of the plurality of storage resources based on the expected retrieval performance. The method then continues by transmitting a read request for one or more encoded data slices associated with the encoded data slice identifier to the storage resource.

    Resolving write conflicts in a dispersed storage network

    公开(公告)号:US11294765B1

    公开(公告)日:2022-04-05

    申请号:US16860883

    申请日:2020-04-28

    Abstract: A method for execution by a dispersed storage and task (DST) execution unit includes receiving a write slice request from a requesting entity, where the write slice request includes a plurality of encoded data slices and a corresponding plurality of slice names. The plurality of slice names are interpreted to produce a slice name range. Whether a write lock conflict exists is determined based on the slice name range. An unfavorable write slice response is issued to the requesting entity when the write lock conflict is determined to exist. When the write lock conflict is determined to not exist, local storage of the plurality of encoded data slices is initiated, a favorable write slice response is issued to the requesting entity, and the processing system indicates that the slice name range is not locked when completing the local storage of the plurality of encoded data slices.

    Metadata Based Listing in a Distributed Storage System

    公开(公告)号:US20220066879A1

    公开(公告)日:2022-03-03

    申请号:US17454125

    申请日:2021-11-09

    Abstract: A method for execution by storage network begins by identifying a subset of encoded data slices stored in a set of storage units for enhanced access and interpreting registry data to determine a listing type. The method continues by generating listing slice names for the subset of encoded data slices based on the listing type and sending the listing slice names to the set of storage units, where each storage unit updating storage and/or listing of at least some of the subset of encoded data slices associated with the storage unit based on the listing type and each storage unit issuing slice access information to a requesting entity in response to one or more slice access requests in accordance with the listing type for some of the subset of encoded data slices and the requesting entity receiving slice access information from at least some of the set of storage units.

    Data restoration using partially encoded slice requests

    公开(公告)号:US11113008B1

    公开(公告)日:2021-09-07

    申请号:US16547809

    申请日:2019-08-22

    Abstract: A storage unit (SU) includes an interface configured to interface and communicate with a dispersed storage network (DSN), a memory that stores operational instructions, and processing circuitry operably coupled to the interface and to the memory. The processing circuitry is configured to execute the operational instructions to perform various operations and functions. The SU, of a first SU set, receives a partially encoded slice request to restore a set of encoded data slices (EDSs) that are based on first dispersed storage error coding function parameters. The SU generates a decode threshold number of partially EDSs based on the first and second dispersed storage error coding function parameters. The SU then outputs the decode threshold number of partially EDSs to a second SU set to undergo selective combination respectively to generate new EDSs for storage within the SUs of the second decode threshold number of SUs of the second SU set.

    Prioritization task execution within a storage unit (SU)

    公开(公告)号:US10831544B1

    公开(公告)日:2020-11-10

    申请号:US16693742

    申请日:2019-11-25

    Abstract: A computing device includes an interface configured to interface and communicate with a dispersed storage network (DSN), a memory that stores operational instructions, and a processing module operably coupled to the interface and memory such that the processing module, when operable within the computing device based on the operational instructions, is configured to perform various operations. For example, the computing device generates a prioritized request that includes at least one of a task for execution or a priority level based on information stored within a storage unit (SU) of a plurality of storage units (SUs) implemented within the DSN. Note that the information corresponds to a data object that is related to a set of encoded data slices (EDSs) that are distributedly stored within the DSN. The computing device then transmits the prioritized request to the SU and receives, from the SU, a response to the prioritized request.

Patent Agency Ranking