Policy-driven raid rebuild

• US 10,642,690 B1
• Filed: 12/20/2017
• Issued: 05/05/2020
• Est. Priority Date: 12/20/2016
• Status: Active Grant

• Current Assignee: Tintri Incorporated
• Original Assignee: Tintri Incorporated
• Inventors: Tian, Lei, Dayal, Shobhit, Gud, Amit
• Primary Examiner(s): Lottich, Joshua P
• Application Number:
  US15/848,330
• Time in Patent Office: 867 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 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0617 in relation to availability
• 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 3/0643 Management of files
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A system for policy-driven RAID rebuild includes an interface to a group of devices each having stripe units. At least one of the devices is a spare device available to be used in the event of a failure of a device in the group of devices. The system further includes a processor coupled to the interface and configured to determine, based at least in part on an ordering criteria, an order in which to reconstruct stripe units to rebuild a failed device in the group of devices. The processor is further configured to rebuild the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed.

Claims

1. A system comprising:

an interface to a group of devices each having stripe units, wherein at least one of the devices is a spare device available to be used in the event of a failure of a device in the group of devices, the group of devices services a Log-structured File System (LFS), and an ordering criteria includes a quantity of valid data and probability of overwrite of in-use segments; and

a processor coupled to the interface, the processor configured to;

determine, based at least in part on the ordering criteria, an order in which to reconstruct stripe units to rebuild a failed device in the group of devices; and

rebuild the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed.

2. The system of claim 1, wherein the group of devices comprise a redundant array of independent disks (RAID) group.

3. The system of claim 1, wherein the group of devices stores data according to one of RAID 5 protocol and RAID 6 protocol.

4. The system of claim 1, wherein the ordering criteria includes workload dynamics in the group of devices.

5. The system of claim 1, wherein the order criteria includes a performance specification indicated by a user.

6. The system of claim 1, wherein the order criteria includes a weighted combination of a plurality of policies.

7. The system of claim 6, wherein the weighted combination of a plurality of policies is updated dynamically.

8. The system of claim 6, wherein the plurality of policies includes at least one of:

rebuild time, user performance, and data availability.

9. The system of claim 1, wherein the ordering criteria includes a frequency of access of data associated with a location in the group of devices.

10. The system of claim 1, wherein the ordering criteria includes a frequency of access of a stripe unit.

11. The system of claim 1, wherein the determination of an order includes scoring stripe units with respect to the ordering criteria.

12. The system of claim 11, wherein the scoring includes combining at least one of a rebuild time score, a user performance score, and a data availability score.

13. The system of claim 1, wherein the determined order is based at least in part on a weighted combination of scores.

14. A system comprising:

an interface to a group of devices each having stripe units, wherein at least one of the devices is a spare device available to be used in the event of a failure of a device in the group of devices;

a processor coupled to the interface, the processor configured to;

determine, based at least in part on an ordering criteria, an order in which to reconstruct stripe units to rebuild a failed device in the group of devices; and

rebuild the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed; and

a cache adapted to store at least one of;

reads and writes, and the ordering criteria includes importance of data and recoverability of data based at least in part on content of the cache.

15. The system of claim 14, wherein the importance of data is based at least in part on filesystem metadata.

16. The system of claim 14, wherein the importance of data is based at least in part on a type of data.

17. A system comprising:

an interface to a group of devices each having stripe units, wherein at least one of the devices is a spare device available to be used in the event of a failure of a device in the group of devices; and

a processor coupled to the interface, the processor configured to;

determine, based at least in part on an ordering criteria, an order in which to reconstruct stripe units to rebuild a failed device in the group of devices; and

rebuild the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed, wherein a rebuild locality score is based at least in part on likelihood of overwriting a stripe unit as indicated by a ranking of an in-use segment in a LFS in-use segment list.

18. The system of claim 17, wherein the rebuild locality score corresponds to a locality of data to be rebuilt and indicates how long it takes to rebuild a stripe unit at the corresponding locality.

19. The system of claim 17, wherein the rebuild locality score prioritizes those stripe units closer to a stripe currently being rebuilt over those strip units farther away from a stripe currently being rebuilt.

20. The system of claim 17, wherein the rebuild locality score prioritizes those stripe units with a lower likelihood of overwriting for rebuilding earlier over those strip units with a higher likelihood of overwriting.

21. A method comprising:

determining, based at least in part on an ordering criteria including a quantity of valid data and probability of overwrite of in-use segments, an order in which to reconstruct stripe units to rebuild a failed device in a group of devices, wherein the group of devices each has stripe units, at least one of the devices is a spare device available to be used in the event of a failure of a device in the group of devices, and the group of devices services a Log-structured File System (LFS); and

rebuilding the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed.

22. A computer program product embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

determining, based at least in part on an ordering criteria including a quantity of valid data and probability of overwrite of in-use segments, an order in which to reconstruct stripe units to rebuild a failed device in a group of devices, wherein the group of devices each has stripe units, at least one of the devices in the group of devices is a spare device available to be used in the event of a failure of a device in the group of devices, and the group of devices services a Log-structured File System (LFS); and

rebuilding the failed device including by reconstructing stripe units in the determined order using the spare device to overwrite stripe units as needed.