Storing data in accordance with encoded data slice revision levels in a dispersed storage network

• US 10,146,620 B2
• Filed: 11/07/2016
• Issued: 12/04/2018
• Est. Priority Date: 04/26/2010
• Status: Expired due to Fees

• Current Assignee: Pure Storage, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Baptist, Andrew, Leggette, Wesley, Resch, Jason K.
• Primary Examiner(s): Shyu, Jing-Yih
• Application Number:
  US15/345,309
• Publication Number:
  US 20170091034A1
• Time in Patent Office: 757 Days
• 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]
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04L 69/26 Special purpose or proprietary protocols or architectures network applications for proprietary or special purpose networking environments H04L67/12

Abstract

A method includes a computing device sending write request messages to storage units. For instance, a first write request that includes a first group of slice payloads for first encoded data slices and their corresponding revision levels is sent to a first storage unit. The method further includes the first storage unit processing the first write request to include a group of status messages for the first encoded data slices regard revision level status. The method further includes the computing device processing the first write response message to flag the encoded data slice having a revision issue for a rollback message and flag encoded data slices not having a revision issue for a write commit message.

Claims

1. A method comprises:

sending, by a computing device of a dispersed storage network (DSN), a set of write request messages to a set of storage units of the DSN regarding a plurality of sets of encoded data slices, wherein a first write request of the set of write requests includes a first group of slice payloads for first encoded data slices of each set of the plurality of sets of encoded data slices and a corresponding revision level of each of the first encoded data slices;

processing, by a first storage unit of the set of storage units, the first write request by;

temporarily storing the first encoded data slices;

determining whether the corresponding revision level of each of the first encoded data slices is a next revision level in accordance with a current revision level maintained by the first storage unit;

generating a group of status messages for the first encoded data slices based on the determining whether the corresponding revision level of each of the first encoded data slices is the next revision level in accordance with the current revision level maintained by the first storage unit; and

sending, to the computing device, a first write response message that includes the group of status messages;

processing, by the computing device, the first write response message by;

interpreting the group of status messages to determine whether an encoded data slice of the first encoded data slices has a revision level error;

when the encoded data slice of the first encoded data slices has a revision level error, flagging the encoded data slices for a rollback message; and

flagging each encoded data slice of the first encoded data slices not having the revision level error for a write commit message.

2. The method of claim 1 further comprises:

generating, by the computing device, the first write request to include a protocol header and a payload section, wherein the payload section includes a transaction number associated with the first group of slice payloads, wherein a slice payload of the first group of slice payloads further includes one or more of a slice name field and a slice length field.

3. The method of claim 1, wherein the determining whether the corresponding revision level of each of the first encoded data slices is the next revision level further comprises one of:

when the corresponding revision level of one of the first encoded data slices is indicative of a first writing of the one of the first encoded data slices, verifying that a previous version of the one of the first encoded data slices is not presently stored by the first storage unit; and

when the corresponding revision level of the one of the first encoded data slices is indicative of revising the one of the first encoded data slices, verifying that the corresponding revision level is a desired increment of a revision level of an immediately previous version of the one of the first encoded data slices stored by the first storage unit.

4. The method of claim 1 further comprises:

wherein a second write request of the set of write requests includes a second group of slice payloads for second encoded data slices of each set of the plurality of sets of encoded data slices and a corresponding revision level of each of the second encoded data slices;

processing, by a second storage unit of the set of storage units, the second write request by;

temporarily storing the second encoded data slices;

determining whether the corresponding revision level of each of the second encoded data slices is a next revision level in accordance with a current revision level maintained by the second storage unit;

generating a second group of status messages for the second encoded data slices based on the determining whether the corresponding revision level of each of the second encoded data slices is the next revision level in accordance with the current revision level maintained by the second storage unit; and

sending, to the computing device, a second write response message that includes the second group of status messages;

processing, by the computing device, the second write response message by;

interpreting the second group of status messages to determine whether a second encoded data slice of the second encoded data slices has a second revision level error;

when the second encoded data slice of the second encoded data slices has the second revision level error, flagging the second encoded data slices for a second rollback message; and

flagging each second encoded data slice of the second encoded data slices not having the second revision level error for a second write commit message.

5. The method of claim 1 further comprises:

generating a status message of the group of status messages to include one or more of;

operation succeeded message;

a transaction conflict due to another transaction locking access to an encoded data slice of the first encoded data slices stored by the first storage unit;

an address error;

the revision level error; and

operation is not authorized for the encoded data slice of the first encoded data slices.

6. The method of claim 1 further comprises:

for each set of the plurality of sets of encoded data slices having a write threshold number of encoded data slices not having the revision level error, issuing write commit commands; and

for each set of the plurality of sets of encoded data slices having less than the write threshold number of encoded data slices not having the revision level error, issuing rollback commands.

7. A non-transitory computer readable memory comprises:

a first memory section that stores operation instructions that, when executed by a computing device of a dispersed storage network (DSN), causes the computing device to;

send a set of write request messages to a set of storage units of the DSN regarding a plurality of sets of encoded data slices, wherein a first write request of the set of write requests includes a first group of slice payloads for first encoded data slices of each set of the plurality of sets of encoded data slices and a corresponding revision level of each of the first encoded data slices;

a second memory section that stores operation instructions that, when executed by a first storage unit of the set of storage units, causes the first storage unit to;

process the first write request by;

temporarily storing the first encoded data slices;

determining whether the corresponding revision level of each of the first encoded data slices is a next revision level in accordance with a current revision level maintained by the first storage unit;

generating a group of status messages for the first encoded data slices based on the determining whether the corresponding revision level of each of the first encoded data slices is the next revision level in accordance with the current revision level maintained by the first storage unit; and

sending, to the computing device, a first write response message that includes the group of status messages; and

a third memory section that stores operation instructions that, when executed by the computing device, causes the computing device to;

process the first write response message by;

interpreting the group of status messages to determine whether an encoded data slice of the first encoded data slices has a revision level error;

when the encoded data slice of the first encoded data slices has a revision level error, flagging the encoded data slices for a rollback message; and

flagging each encoded data slice of the first encoded data slices not having the revision level error for a write commit message.

8. The non-transitory computer readable memory of claim 7, wherein the first memory section further stores operation instructions that, when executed by the computing device, causes the computing device to:

generate the first write request to include a protocol header and a payload section, wherein the payload section includes a transaction number associated with the first group of slice payloads, wherein a slice payload of the first group of slice payloads further includes one or more of a slice name field and a slice length field.

9. The non-transitory computer readable memory of claim 7, wherein the second memory section further stores operation instructions that, when executed by the first storage unit, causes the first storage unit to determine whether the corresponding revision level of each of the first encoded data slices is the next revision level further by one of:

when the corresponding revision level of one of the first encoded data slices is indicative of a first writing of the one of the first encoded data slices, verifying that a previous version of the one of the first encoded data slices is not presently stored by the first storage unit; and

when the corresponding revision level of the one of the first encoded data slices is indicative of revising the one of the first encoded data slices, verifying that the corresponding revision level is a desired increment of a revision level of an immediately previous version of the one of the first encoded data slices stored by the first storage unit.

10. The non-transitory computer readable memory of claim 7 further comprises:

wherein a second write request of the set of write requests includes a second group of slice payloads for second encoded data slices of each set of the plurality of sets of encoded data slices and a corresponding revision level of each of the second encoded data slices;

a fourth memory section that stores operation instructions that, when executed by a second storage unit of the set of storage units, causes the second storage unit to;

process the second write request by;

temporarily storing the second encoded data slices;

determining whether the corresponding revision level of each of the second encoded data slices is a next revision level in accordance with a current revision level maintained by the second storage unit;

generating a second group of status messages for the second encoded data slices based on the determining whether the corresponding revision level of each of the second encoded data slices is the next revision level in accordance with the current revision level maintained by the second storage unit; and

sending, to the computing device, a second write response message that includes the second group of status messages;

the third memory section further stores operation instructions that, when executed by the computing device, causes the computing device to;

process the second write response message by;

interpreting the second group of status messages to determine whether a second encoded data slice of the second encoded data slices has a second revision level error;

when the second encoded data slice of the second encoded data slices has the second revision level error, flagging the second encoded data slices for a second rollback message; and

flagging each second encoded data slice of the second encoded data slices not having the second revision level error for a second write commit message.

11. The non-transitory computer readable memory of claim 7, wherein the second memory section further stores operation instructions that, when executed by the first storage unit, causes the first storage unit to:

generate a status message of the group of status messages to include one or more of;

operation succeeded message;

a transaction conflict due to another transaction locking access to an encoded data slice of the first encoded data slices stored by the first storage unit;

an address error;

the revision level error; and

operation is not authorized for the encoded data slice of the first encoded data slices.

12. The non-transitory computer readable memory of claim 7, wherein the third memory section further stores operation instructions that, when executed by the computing device, causes the computing device to:

for each set of the plurality of sets of encoded data slices having a write threshold number of encoded data slices not having the revision level error, issue write commit commands; and

for each set of the plurality of sets of encoded data slices having less than the write threshold number of encoded data slices not having the revision level error, issue rollback commands.