International Business Machines Corporation

Pure Storage, Inc.

Renars W Narubin

Jason K Resch

Manish Motwani

Ravi V Khadiwala

Yogesh Ramesh Vedpathak

Daniel J Scholl

Gary W Grube

Thomas Dubucq

Jeremy S Jarczyk

Greg R Dhuse

Andrew D Baptist

Ilya Volvovski

Joseph M Kaczmarek

A method includes error encoding data to produce a plurality of data slices. Metadata is determined for a data slice of the plurality of data slices. The metadata is stored in a metadata storage tree. The data slice is stored in a slice storage location indicated by the metadata. Based on determining to access the data slice, the metadata for the data slice is accessed in the metadata storage tree to determine the slice storage location for the data slice, and the data slice is accessed in the slice storage location based on determining the slice storage location for the data slice via accessing the metadata storage tree.

Accessing metadata storage trees in a storage network

A method includes receiving data for storage and encoding the data to produce a plurality of data slices. Metadata is determined for a data slice of the plurality of data slices. The metadata is stored in a metadata storage tree. The metadata storage tree is stored via a first plurality of memory devices of a first memory type. The data slice is stored in a slice storage location in a second plurality of memory devices of a second memory type. The slice storage location is indicated by the metadata. The first memory type has a higher performance level than the second memory type based on a utilization approach.

Utilizing metadata storage trees in a vast storage network

A method for execution by a dispersed storage (DS) cleanup unit includes determining a dead session of a DSN. A subset of a plurality of eventual consistency intent names is generated by identifying eventual consistency intent names that include a session identifier corresponding to the dead session in a prefix of the eventual consistency intent names, where the subset of the plurality of eventual consistency intent names corresponds to all eventual consistency intents of the dead session. A subset of storage units responsible for storing the all eventual consistency intents of the dead session is determined based on the prefix of the eventual consistency intent names in the subset. All eventual consistency intents of the dead session are retrieved from the subset of storage units, and execution of eventual consistency updates indicated in the all eventual consistency intents of the dead session is facilitated.

Eventual consistency intent cleanup in a dispersed storage network

A method includes a storage unit (SU) of a dispersed storage network (DSN) receiving and storing a first modification for an encoded data slice (EDS) associated with the SU and assigning an identifier for the first modification. The method continues with the SU receiving and storing a second modification for the EDS and assigning an identifier for the second modification. The method continues with the SU receiving a request to relocate the EDS to another SU of the DSN. The method continues with the SU transmitting the EDS to the another SU of the DSN, followed by transmitting the first modification for the EDS and the identifier to the another SU of the DSN in a first message. The method continues with the SU transmitting the second modification with the identifier for the second modification to the another SU of the DSN in a second message, where the second message is separate from the first message.

Stateful relocator for a distributed storage network

A method for execution by a dispersed storage and task (DST) execution unit includes receiving a slice write request via a network that includes a data slice and extracting metadata from the data slice. The metadata is stored in a metadata storage tree in a first memory device of the DST execution unit and the data slice is stored in a slice storage tree in a second memory device of the DST execution unit based on tree utilization parameters.

Utilizing tree storage structures in a dispersed storage network

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 a processing module operably coupled to the interface and memory such that the processing module, when operable within the SU based on the operational instructions, is configured to perform various operations. The SU stores at least one encoded data slice (EDS) of first EDSs corresponding to a data object that are distributedly stored in first SUs and also an intent message that includes specifications for consistency between the data object and metadata of the data object. A second set of EDSs corresponding to the metadata are distributedly stored in second SUs. The SU services the intent message to determine consistency of the data object and the metadata based on the specifications and deletes the intent message when they are consistent.

Ensuring consistency between content and metadata with intents

A method for execution by a dispersed storage and task (DST) processing unit includes generating a plurality of access requests that include an execution deadline time for transmission via a network to a corresponding subset of a plurality of storage units. A first deadline error notification is received via the network from a first storage unit of the first subset. A new one of the plurality of storage units not included in the first subset is selected in response to receiving the first deadline error notification. A new access request that includes an updated execution deadline time is generated for transmission to the new one of the plurality of storage units via the network. The new access request is based on a one of the first plurality of access requests sent to the first storage unit of the first subset.

Utilizing request deadlines in a dispersed storage network

A method beings by a computing device receiving a write request for a data segment that has been encoded to produce a set of encoded data slices. The method continues with the write request being stored in memory and a write intent associated with the write request being created and stored as an object in memory. The computing device then determines whether metadata associated with the data segment can be updated, and when the metadata cannot be updated maintaining the write request in memory until a cleanup agent can execute the write intent and successfully update the metadata.

Growing community of inventors

Chicago, IL, United States of America

Renars W Narubin