Directory synchronization of a dispersed storage network

• US 9,501,341 B2
• Filed: 09/09/2014
• Issued: 11/22/2016
• Est. Priority Date: 03/31/2008
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Dhuse, Greg, Baptist, Andrew, Leggette, Wesley, Volvovski, Ilya, Resch, Jason K., Cilfone, Bart
• Primary Examiner(s): Desrosiers, Evans
• Application Number:
  US14/481,892
• Publication Number:
  US 20140379675A1
• Time in Patent Office: 805 Days
• Field of Search: 713/193, 713/189, 709220-229, 709/248, 726 1- 36, 380/229
• US Class Current: 1/1
• CPC Class Codes: G06F 11/0745 in an input/output transactions management context input/output processing in general G06F13/00
• CPC Class Codes: G06F 11/1064 in cache or content addressable memories
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/1443 Transmit or communication errors
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 12/0813 with a network or matrix configuration
• CPC Class Codes: G06F 12/0815 Cache consistency protocols
• CPC Class Codes: G06F 21/10 Protecting distributed programs or content, e.g. vending or licensing of copyrighted material protection in video systems or pay television H04N7/16 ; Digital rights management [DRM]
• CPC Class Codes: G06F 2211/1028 Distributed, i.e. distributed RAID systems with parity
• CPC Class Codes: G06F 2211/1057 Parity-multiple bits-RAID6, i.e. RAID 6 implementations
• CPC Class Codes: G06F 9/467 Transactional memory G06F9/528 takes precedence

Abstract

A method begins by independently executing a first write transaction in a dispersed storage network (DSN) to a particular write verification step of a multiple step write process, wherein the first write transaction has a first transaction identifier. The method continues by independently executing a second write transaction in the DSN to the particular write verification step, wherein the second write transaction has a second transaction identifier, and wherein subject matter of the first write transaction is related to subject matter of the second write transaction. The method continues by dependently finalizing the multiple step write process for each of the first and second write transactions utilizing the first and second transaction identifiers when each of the first and second write transactions have reached the particular write verification step.

Claims

1. A method for execution by a computing device, the method comprises:

independently executing a first write transaction in a dispersed storage network (DSN) to a particular write verification step of a multiple step write process, wherein the first write transaction has a first transaction identifier, the first write transaction includes a first set of write commands for a data segment that has been dispersed storage error encoded into a set of encoded data slices;

independently executing a second write transaction in the DSN to the particular write verification step, wherein the second write transaction has a second transaction identifier, the second write transaction includes a second set of write commands for directory information regarding the set of encoded data slices, wherein the directory information is dispersed storage error encoded into a set of encoded directory slices and wherein subject matter of the first write transaction is related to subject matter of the second write transaction; and

when each of the first and second write transactions have reached the particular write verification step, dependently finalizing the multiple step write process for each of the first and second write transactions utilizing the first and second transaction identifiers.

2. The method of claim 1, wherein the multiple step write process further comprises:

a write initiate step that includes a set of write requests and a write threshold number of corresponding write responses;

a write commit step that includes a set of write commit requests and the write threshold number of corresponding write commit responses; and

a write finalize step that includes a set of write finalize commands, wherein the particular step is the write initiate step.

3. The method of claim 1, wherein the multiple step write process further comprises:

a write initiate step that includes a set of write requests and a write threshold number of corresponding write responses;

a write commit step that includes a set of write commit requests and the write threshold number of corresponding write commit responses; and

a write finalize step that includes a set of write finalize commands, wherein the particular step is the write commit step.

4. The method of claim 1, wherein the independently executing the first write transaction further comprises:

generating the first transaction identifier regarding storage of a set of encoded data slices, wherein a data segment was dispersed storage error encoded to produce the set of encoded data slices;

outputting a set of write request messages to dispersed storage (DS) units of the DSN, wherein each write request message of the set of write request messages includes the first transaction identifier and a corresponding one of the set of encoded data slices;

receiving write response messages from at least some of the DS units, wherein each of the write response messages includes a reference to the first transaction identifier; and

when at least a write threshold number of the write response messages have been received, commencing a commit step of the multiple step write process for the first write transaction.

5. The method of claim 4, wherein the commencing the commit step for the first write transaction comprises:

outputting a plurality of data commit request messages regarding the set of encoded data slices to the DS units, wherein each of the plurality of data commit request messages includes the first transaction identifier.

6. The method of claim 1, wherein the independently executing the second write transaction further comprises:

generating the second transaction identifier regarding storage of a set of encoded directory slices, wherein directory information is dispersed storage error encoded to produce a set of encoded directory slices;

outputting a second plurality of write request messages to DS units, wherein each of the second plurality of write request messages includes the second transaction identifier and a corresponding one of the set of encoded directory slices; and

when at least a second write threshold number of favorable second write response messages have been received from the DS units, commencing a commit step of the multiple step write process for the second write transaction.

7. The method of claim 6, wherein the commencing the commit step for the second write transaction comprises:

outputting a plurality of directory commit request messages regarding the set of encoded directory slices to the DS units, wherein each of the plurality of directory commit request messages includes the second transaction identifier.

8. A computer comprises:

an interface;

memory; and

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

independently execute a first write transaction in a dispersed storage network (DSN) to a particular write verification step of a multiple step write process, wherein the first write transaction has a first transaction identifier, the first write transaction including a first set of write commands for a data segment that has been dispersed storage error encoded into a set of encoded data slices;

independently execute a second write transaction in the DSN to the particular write verification step, wherein the second write transaction has a second transaction identifier, the second write transaction including a second set of write commands for directory information regarding the set of encoded data slices, wherein the directory information is dispersed storage error encoded into a set of encoded directory slices, and wherein subject matter of the first write transaction is related to subject matter of the second write transaction; and

when each of the first and second write transactions have reached the particular write verification step, dependently finalize the multiple step write process for each of the first and second write transactions utilizing the first and second transaction identifiers.

9. The computer of claim 8, wherein the processing module is further operable to:

output, via the interface, a write initiate step that includes a set of write requests and a write threshold number of corresponding write responses;

output, via the interface, a write commit step that includes a set of write commit requests and the write threshold number of corresponding write commit responses; and

output, via the interface, a write finalize step that includes a set of write finalize commands, wherein the particular step is the write initiate step.

10. The computer of claim 8, wherein the processing module is further operable to:

output, via the interface, a write initiate step that includes a set of write requests and a write threshold number of corresponding write responses;

output, via the interface, a write commit step that includes a set of write commit requests and the write threshold number of corresponding write commit responses; and

output, via the interface, a write finalize step that includes a set of write finalize commands, wherein the particular step is the write commit step.

11. The computer of claim 8, wherein the processing module is further operable to:

generate the first transaction identifier regarding storage of a set of encoded data slices, wherein a data segment was dispersed storage error encoded to produce the set of encoded data slices;

output, via the interface, a set of write request messages to dispersed storage (DS) units of the DSN, wherein each write request message of the set of write request messages includes the first transaction identifier and a corresponding one of the set of encoded data slices;

receive, via the interface, write response messages from at least some of the DS units, wherein each of the write response messages includes a reference to the first transaction identifier; and

when at least a write threshold number of the write response messages have been received, commence a commit step of the multiple step write process for the first write transaction.

12. The computer of claim 11, wherein the processing module further functions to commence the commit step for the first write transaction by:

outputting, via the interface, a plurality of data commit request messages regarding the set of encoded data slices to the DS units, wherein each of the plurality of data commit request messages includes the first transaction identifier.

13. The computer of claim 8, wherein the processing module further functions to independently execute the second write transaction by:

generating the second transaction identifier regarding storage of a set of encoded directory slices, wherein directory information is dispersed storage error encoded to produce a set of encoded directory slices;

outputting, via the interface, a second plurality of write request messages to DS units, wherein each of the second plurality of write request messages includes the second transaction identifier and a corresponding one of the set of encoded directory slices; and

when at least a second write threshold number of favorable second write response messages have been received from the DS units, commencing a commit step of the multiple step write process for the second write transaction.

14. The computer of claim 13, wherein the processing module further functions to commence the commit step for the second write transaction by:

outputting, via the interface, a plurality of directory commit request messages regarding the set of encoded directory slices to the DS units, wherein each of the plurality of directory commit request messages includes the second transaction identifier.

15. A method for execution by a computing device, the method comprises:

independently executing a first write transaction in a dispersed storage network (DSN) to a particular write verification step of a multiple step write process, wherein the first write transaction has a first transaction identifier, the independently executing the first write transaction including;

generating the first transaction identifier regarding storage of a set of encoded data slices, wherein a data segment was dispersed storage error encoded to produce the set of encoded data slices;

outputting a set of write request messages to dispersed storage (DS) units of the DSN, wherein each write request message of the set of write request messages includes the first transaction identifier and a corresponding one of the set of encoded data slices;

receiving write response messages from at least some of the DS units, wherein each of the write response messages includes a reference to the first transaction identifier; and

when at least a write threshold number of the write response messages have been received, commencing a commit step of the multiple step write process for the first write transaction; and

independently executing a second write transaction in the DSN to the particular write verification step, wherein the second write transaction has a second transaction identifier, and wherein subject matter of the first write transaction is related to subject matter of the second write transaction; and

when each of the first and second write transactions have reached the particular write verification step, dependently finalizing the multiple step write process for each of the first and second write transactions utilizing the first and second transaction identifiers.

16. The method of claim 15, wherein the commencing the commit step for the first write transaction comprises:

outputting a plurality of data commit request messages regarding the set of encoded data slices to the DS units, wherein each of the plurality of data commit request messages includes the first transaction identifier.

17. The method of claim 15, wherein the independently executing the second write transaction further comprises:

generating the second transaction identifier regarding storage of a set of encoded directory slices, wherein directory information is dispersed storage error encoded to produce a set of encoded directory slices;

outputting a second plurality of write request messages to DS units, wherein each of the second plurality of write request messages includes the second transaction identifier and a corresponding one of the set of encoded directory slices; and

when at least a second write threshold number of favorable second write response messages have been received from the DS units, commencing a commit step of the multiple step write process for the second write transaction.

18. The method of claim 17, wherein the commencing the commit step for the second write transaction comprises:

outputting a plurality of directory commit request messages regarding the set of encoded directory slices to the DS units, wherein each of the plurality of directory commit request messages includes the second transaction identifier.