Read-source requests to support bundled writes in a distributed storage system

• US 10,423,490 B2
• Filed: 12/06/2017
• Issued: 09/24/2019
• Est. Priority Date: 01/30/2015
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Baptist, Andrew D., Dhuse, Greg R., Khadiwala, Ravi V., Resch, Jason K., Leggette, Wesley B., Vossberg, Trevor J.
• Primary Examiner(s): Alphonse, Fritz
• Application Number:
  US15/833,760
• Publication Number:
  US 20180095827A1
• Time in Patent Office: 657 Days
• Field of Search: 714764, 714768, 714765
• CPC Class Codes: G06F 11/0739 in a data processing system embedded in automotive or aircraft systems
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A method for execution by processing modules of one or more computing devices of a dispersed storage network (DSN), the method begins by identifying a stored data object (using bundled writes) for retrieval from a dispersed storage network (DSN), determining a DSN address that corresponds to the store data object, generating a read source request based on the DSN address, identifying a set of storage units of the DSN, where one or more of the storage units of the set of storage units are associated with storage of the stored data object, sending the read source request to the identified set of storage units, receiving retrieved encoded data slices from at least some of the storage units of the identified set of storage units and dispersed storage error decoding, for each set of encoded data slices, a decode threshold of received encoded data slices to produce a recovered data object.

Claims

1. A method for execution by one or more processing modules of one or more computing devices of a dispersed storage network (DSN), the method comprises:

identifying a stored data object for retrieval from a dispersed storage network (DSN), wherein the stored data object is stored at least partially as a bundled write;

determining a DSN address that corresponds to the stored data object;

generating a read source request based on the DSN address, wherein the read source request includes a common source name;

identifying a set of storage units of the DSN, wherein one or more storage units of the set of storage units are associated with the common source name;

sending the read source request to the identified set of storage units to retrieve encoded data slices associated with the common source name;

receiving the retrieved encoded data slices from at least some of the storage units of the identified set of storage units, wherein each storage unit that receives the read source request returns all stored encoded data slices with a source name that matches the common source name provided in the read source request; and

disperse storage error decoding, for each set of the encoded data slices, a decode threshold number of received encoded data slices to produce a recovered data object.

2. The method of claim 1, wherein the identifying a stored data object for retrieval includes at least one of interpreting a request or performing a lookup.

3. The method of claim 1, wherein the determining a DSN address includes one or more of interpreting a dispersed hierarchical index, interpreting a DSN directory, or identifying a source name common to all slice names of encoded data slices of the store data object.

4. The method of claim 1, wherein the generating the read source request includes a source name that is common to all slice names of encoded data slices of the store data object.

5. The method of claim 1, wherein the identifying a set of storage units of the DSN includes one or more of issuing a query, interpreting a query response, interpreting an error message, interpreting a storage unit status, or performing a lookup.

6. The method of claim 1, wherein sending the read source request to the identified set of storage units includes replicating the read source request and transmitting a replicated resource request to each storage unit of the identified set of storage units.

7. The method of claim 1, wherein the receiving the retrieved encoded data slices includes receiving read slice responses and extracting one or more retrieved encoded data slices from the received read slice responses.

8. The method of claim 1, wherein the dispersed storage error decoding, for each set of encoded data slices, includes identifying a decode threshold number of received encoded data slices, dispersed storage error decoding the decode threshold number of received encoded data slices to reproduce a data segment of a plurality of data segments of the stored data object, and aggregating the received encoded data slices to produce the recovered data object.

9. A computing device of a group of computing devices of a dispersed storage network (DSN), the computing device comprises:

an interface;

a local memory; and

a processing module operably coupled to the interface and the local memory, wherein the processing module is configured to;

identify a stored data object, stored as encoded data slices, for retrieval from a dispersed storage network (DSN), wherein the stored data object is stored at least partially as a bundled write;

determine a DSN address that corresponds to the stored data object;

generate a read source request based on the DSN address, wherein the read source request includes a common source name, wherein the generate a read source request includes a source name that is common to all slice names of the encoded data slices of the stored data object;

identify a set of storage units of the DSN, wherein one or more storage units of the set of storage units are associated with the common source name;

send the read source request to the identified set of storage units to retrieve encoded data slices associated with the common source name;

receive the retrieved encoded data slices from at least some of the storage units of the identified set of storage units, wherein each storage unit that receives the read source request returns all stored encoded data slices with a source name that matches the common source name provided in the read source request; and

disperse storage error decode, for each set of the encoded data slices, a decode threshold number of received encoded data slices to produce a recovered data object.

10. The computing device of claim 9, wherein the identify a stored data object for retrieval includes at least one of interpreting a request or performing a lookup.

11. The computing device of claim 9, wherein the determine a DSN address includes one or more of interpreting a dispersed hierarchical index, interpreting a DSN directory, or identifying a source name common to all slice names of the encoded data slices of the stored data object.

12. The computing device of claim 9, wherein the identify a set of storage units of the DSN includes one or more of issuing a query, interpreting a query response, interpreting an error message, interpreting a storage unit status, or performing a lookup.

13. The computing device of claim 9, wherein the send the read source request to the identified set of storage units includes replicating the read source request and transmitting a replicated resource request to each storage unit of the identified set of storage units.

14. The computing device of claim 9, wherein the receive the retrieved encoded data slices includes receiving read slice responses and extracting one or more retrieved encoded data slices from the received read slice responses.

15. The computing device of claim 9, wherein the disperse storage error decode, for each set of the encoded data slices, includes identifying a decode threshold number of received encoded data slices, dispersed storage error decoding the decode threshold number of received encoded data slices to reproduce a data segment of a plurality of data segments of the stored data object, and aggregating the received encoded data slices to produce the recovered data object.

16. A system, comprises:

an interface;

a local memory; and

a processing module operably coupled to the interface and the local memory, wherein the processing module is configured to;

identify a stored data object for retrieval from a dispersed storage network (DSN), wherein the stored data object is stored at least partially as a bundled write;

determine a DSN address that corresponds to the stored data object;

generate a read source request based on the DSN address, wherein the read source request includes a common source name;

identify a set of storage units of the DSN, wherein one or more storage units of the set of storage units are associated with the common source name;

send the read source request to the identified set of storage units to retrieve encoded data slices associated with the common source name, wherein the send the read source request to the identified set of storage units includes replicating the read source request and transmitting a replicated resource request to each storage unit of the identified set of storage units;

receive the retrieved encoded data slices from at least some of the storage units of the identified set of storage units, wherein each storage unit that receives the read source request returns all stored encoded data slices with a source name that matches the common source name provided in the read source request; and

disperse storage error decode, for each set of the encoded data slices, a decode threshold number of received encoded data slices to produce a recovered data object.

17. The system of claim 16, wherein the determine a DSN address includes one or more of interpreting a dispersed hierarchical index, interpreting a DSN directory, or identifying a source name common to all slice names of encoded data slices of the stored data object.

18. The system of claim 16, wherein the generate a read source request includes a source name that is common to all slice names of encoded data slices of the stored data object.