Distributed storage network for modification of a data object

• US 9,098,376 B2
• Filed: 05/30/2014
• Issued: 08/04/2015
• Est. Priority Date: 10/30/2009
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Cleversafe Incorporated (International Business Machines Corporation)
• Inventors: Resch, Jason K.
• Primary Examiner(s): Merant, Guerrier
• Application Number:
  US14/291,205
• Publication Number:
  US 20140281804A1
• Time in Patent Office: 431 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system
• CPC Class Codes: G06F 2211/1028 Distributed, i.e. distributed RAID systems with parity
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57
• CPC Class Codes: H04L 67/06 specially adapted for file transfer, e.g. file transfer protocol [FTP]
• 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 69/14 Multichannel or multilink protocols

Abstract

In a dispersed storage network, data objects are dispersed storage error encoded into pluralities of sets of encoded data slices that are each stored in corresponding sets of storage units. To recover a data object, a read threshold number of encoded data slices from each set of encoded data slices of the plurality of sets of encoded data slices are required. Upon determining that an update is available for the storage units, a dispersed storage managing unit determines an updating sequence pattern that ensures that while one or more storage units are taken off line for the update, a sufficient number of storage units remain on line such that at least the read threshold number of encoded data slices are available for each set of encoded data slices.

Claims

1. A dispersed storage managing unit for use within a dispersed storage network, comprising:

an interface;

memory; and

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

determine that an update is available for a plurality of storage units, wherein sets of storage units of the plurality of storage units support a plurality of virtual memory vaults, wherein a virtual memory vault of the plurality of virtual memory vaults is mapped to a corresponding set of storage units of the sets of storage units, wherein data objects are dispersed storage error encoded into pluralities of sets of encoded data slices that are each stored in the corresponding set of storage units and wherein, to recover a data object of the data objects, a read threshold number of encoded data slices from each set of encoded data slices of the pluralities of sets of encoded data slices are required, and wherein a storage unit of the plurality of storage units supports a set of virtual memory vaults of the plurality of virtual memory vaults;

determine an updating sequencing pattern, wherein the updating sequencing pattern ensures that, while one or more storage units of the plurality of storage units is taken off line for the update, a sufficient number of storage units remain on line such that at least the read threshold number of encoded data slices are available for each set of the pluralities of sets of encoded data slices;

take a first subset of storage units of the plurality of storage units off line in accordance with the updating sequencing pattern to perform the update; and

when the update is completed in the first subset of storage units, bring the first subset of storage units back on line.

2. The dispersed storage managing unit of claim 1, wherein the determining the updated sequencing pattern comprises:

ascertaining a respective read threshold number for each virtual memory vault of the plurality of virtual memory vaults to produce a set of read threshold numbers;

for each of the plurality of storage units, identifying one or more of the plurality of virtual memory vaults that the storage unit supports to produce a plurality of lists of supported virtual memory vaults; and

utilizing the set of read threshold numbers and the plurality of lists of supported virtual memory vaults to produce the updating sequencing pattern.

3. The dispersed storage managing unit of claim 1, further comprises:

the update being a software update; and

the processing module is further operable to;

when the update is complete in the first subset of storage units, bring the first subset of storage units back on line with the software update disabled until at least the read threshold number of storage units in a first set of storage units associated with a first virtual memory vault of the plurality of virtual memory vaults have successfully received the software update.

4. The dispersed storage managing unit of claim 3, wherein the processing module is further operable to:

track a number of storage units of the first set of storage units that have successfully received the software update; and

when the number of storage units that have successfully received the software update is equal to or greater than the read threshold number of storage units for the first virtual memory vault, send a signal to the number of storage units that have successfully received the software update to enable use of the software update.

5. The dispersed storage unit of claim 3, wherein the processing module is further operable to:

download the software update via the interface; and

cache the software update in the memory.

6. The dispersed storage managing unit of claim 3, wherein the processing module is further operable to perform at least one of:

provide the software update to the first subset of storage units; and

provide a message that instructs the first subset of storage units to retrieve the software update from a trusted source.

7. The dispersed storage managing unit of claim 3, wherein the processing module is further operable to:

provide the software update to all of the plurality of storage units prior to taking the first subset of storage units off line.

8. The dispersed storage managing unit of claim 3, further comprises:

the memory storing a current software version running on each of the plurality of storage units; and

the processing module is further operable to;

compare an update version of the software update with the current software version to determine whether to provide the software update to the plurality of storage units.

9. The dispersed storage managing unit of claim 3, wherein the processing module is further operable to:

take the first subset of storage units of the first set of storage units off line to perform the software update;

when the software update is complete in the first subset of storage units, bring the first subset of storage units back on line with the software update disabled;

enter a first loop that includes;

taking a next subset of storage units of the first set of storage units off line to perform the software update;

when the software update is complete in the next subset of storage units, bringing the next subset of storage units back on line with the software update disabled;

determining whether the read threshold number of storage units of the first set of storage units have successfully received the software update;

when the read threshold number of storage units have not successfully received the software update, repeating the first loop with another subset of storage units of the first set of storage units as the next subset of storage units;

when the read threshold number of storage units have successfully received the software update, exiting the first loop and sending a message to the read threshold number of storage units in the first set of storage units to enable the software update; and

enter a second loop that includes;

taking a subsequent subset of storage units of the first set of storage units off line to perform the software update;

when the software update is complete in the subsequent subset of storage units, bringing the subsequent subset of storage units back on line with the software update enabled;

determining whether the subsequent subset of storage units includes the last storage unit of the first set of storage units;

when the subsequent subset of storage units does not include the last storage unit of the first set of storage units, repeating the second loop with a further subset of storage units of the first set of storage units as the subsequent subset of storage units; and

when the subsequent subset of storage units includes the last storage unit of the first set of storage units, exiting the second loop.

10. The dispersed storage managing unit of claim 3, further comprises:

a decode threshold number including the read threshold number and a safety factor;

the read threshold number further indicating a minimum number of storage units in the first set of storage units that are needed to enable the retrieval of the data object; and

the safety factor indicating an additional number of storage units in addition to the minimum number of storage units in the first set of storage units that should remain on line during the update;

wherein a remaining number of storage units in the first set of storage units that remain on line while the first subset of storage units is updated is at least equal to the decode threshold number.

11. The dispersed storage managing unit of claim 1, further comprises:

the update being a software update; and

the processing module is further operable to;

when the software update is completed in the first subset of storage units, bring the first subset of storage units back on line with the software update disabled;

enter a loop that includes;

taking a next subset of storage units of the plurality of storage units off line to perform the software update; and

when the software is updated in the next subset of storage units, bringing the next subset of storage units back on line with the software update disabled;

determining whether the next subset of storage units includes the last storage unit of the plurality of storage units;

when the next subset of storage units does not include the last storage unit of the plurality of storage units, repeating the loop with another subset of storage units of the plurality of storage units as the next subset of storage units;

when the next subset of storage units does include the last storage unit of the plurality of storage units, exiting the loop and sending a message to the plurality of storage units to enable the software update.

12. The dispersed storage managing unit of claim 1, wherein the update comprises one or more of:

a new software application;

a new version of an existing software application;

a new version of a portion of an existing software application;

adding new physical memory;

updating physical memory;

adding a new processing module;

updating a processing module;

updating an interface; and

adding a new interface.

13. A method for updating a set of storage units within a dispersed storage network, comprising:

determining, by a processing module, that an update is available for a plurality of storage units, wherein sets of storage units of the plurality of storage units support a plurality of virtual memory vaults, wherein a virtual memory vault of the plurality of virtual memory vaults is mapped to a corresponding set of storage units of the sets of storage units, wherein data objects are dispersed storage error encoded into pluralities of sets of encoded data slices that are each stored in the corresponding set of storage units and wherein, to recover a data object of the data objects, a read threshold number of encoded data slices from each set of encoded data slices of the pluralities of sets of encoded data slices are required, and wherein a storage unit of the plurality of storage units supports a set of virtual memory vaults of the plurality of virtual memory vaults;

determining, by the processing module, an updating sequencing pattern, wherein the updating sequencing pattern ensures that, while one or more storage units of the plurality of storage units is taken off line for the update, a sufficient number of storage units remain on line such that at least the read threshold number of encoded data slices are available for each set of the pluralities of sets of encoded data slices;

taking, by the processing module, a first subset of storage units of the plurality of storage units off line in accordance with the updating sequencing pattern to perform the update; and

when the update is completed in the first subset of storage units, bringing, by the processing module, the first subset of storage units back on line.

14. The method of claim 13, wherein the determining the updated sequencing pattern comprises:

ascertaining, by the processing module, a respective read threshold number for each virtual memory vault of the plurality of virtual memory vaults to produce a set of read threshold numbers;

for each of the plurality of storage units, identifying, by the processing module, one or more of the plurality of virtual memory vaults that the storage unit supports to produce a plurality of lists of supported virtual memory vaults; and

utilizing, by the processing module, the set of read threshold numbers and the plurality of lists of supported virtual memory vaults to produce the update sequencing pattern.

15. The method of claim 13, wherein the update is a software update and further comprises:

when the update is complete in the first subset of storage units, bringing, by the processing module, the first subset of storage units back on line with the software update disabled until at least the read threshold number of storage units in a first set of storage units associated with a first virtual memory vault of the plurality of virtual memory vaults have successfully received the software update.

16. The method of claim 15, further comprises:

tracking, by the processing module, a number of storage units of the first set of storage units that have successfully received the software update; and

when the number of storage units that have successfully received the software update is equal to or greater than the read threshold number of storage units for the first virtual memory vault, sending, by the processing module, a signal to the number of storage units that have successfully received the software update to enable use of the software update.

17. A non-transitory computer readable storage medium having accessible therefrom a set of instructions interpretable by a processing module, the set of instructions being configured to cause the processing module to carry out operations for:

determining that an update is available for a plurality of storage units, wherein sets of storage units of the plurality of storage units support a plurality of virtual memory vaults, wherein a virtual memory vault of the plurality of virtual memory vaults is mapped to a corresponding set of storage units of the sets of storage units, wherein data objects are dispersed storage error encoded into pluralities of sets of encoded data slices that are each stored in the corresponding set of storage units and wherein, to recover a data object of the data objects, a read threshold number of encoded data slices from each set of encoded data slices of the pluralities of sets of encoded data slices are required, and wherein a storage unit of the plurality of storage units supports a set of virtual memory vaults of the plurality of virtual memory vaults;

determining an updating sequencing pattern, wherein the updating sequencing pattern ensures that, while one or more storage units of the plurality of storage units is taken off line for the update, a sufficient number of storage units remain on line such that at least the read threshold number of encoded data slices are available for each set of the pluralities of sets of encoded data slices;

taking a first subset of storage units of the plurality of storage units off line in accordance with the updating sequencing pattern to perform the update; and

when the update is completed in the first subset of storage units, bringing the first subset of storage units back on line.

18. The storage medium of claim 17, wherein the set of instructions further causes the processing module to carry out operations for:

ascertaining a respective read threshold number for each virtual memory vault of the plurality of virtual memory vaults to produce a set of read threshold numbers;

for each of the plurality of storage units, identifying one or more of the plurality of virtual memory vaults that the storage unit supports to produce a plurality of lists of supported virtual memory vaults; and

utilizing the set of read threshold numbers and the plurality of lists of supported virtual memory vaults to produce the update sequencing pattern.

19. The storage medium of claim 17, wherein the update is a software update and the set of instructions further causes the processing module to carry out operations for:

when the update is complete in the first subset of storage units, bringing the first subset of storage units back on line with the software update disabled until at least the read threshold number of storage units in a first set of storage units associated with a first virtual memory vault of the plurality of virtual memory vaults have successfully received the software update.

20. The storage medium of claim 19, wherein the set of instructions further causes the processing module to carry out operations for performing at least one of:

tracking a number of storage units of the first set of storage units that have successfully received the software update; and

when the number of storage units that have successfully received the software update is equal to or greater than the read threshold number of storage units for the first virtual memory vault, sending a signal to the number of storage units that have successfully received the software update to enable use of the software update.