Method and apparatus for redundant array of independent disks

Patent metadata

Field	Value
Publication number	US-2017091052-A1
Application number	US-201615272693-A
Country	US
Kind code	A1
Filing date	Sep 22, 2016
Priority date	Sep 25, 2015
Publication date	Mar 30, 2017
Grant date	—

Abstract

The present disclosure provides a method and apparatus for a redundant array of independent disks, RAID. The method includes: dividing the RAID into a plurality of sub-RAIDs; generating sub-RAID metadata corresponding to each of the sub-RAIDs; and in response to a disk in the RAID being replaced, rebuilding data in a corresponding sub-RAID based on the sub-RAID metadata. The present disclosure further provides a computer readable storage medium and a storage system.

First claim

I/We claim:

1 . A method for a redundant array of independent disks, RAID, comprising:
dividing the RAID into a plurality of sub-RAIDs; generating sub-RAID metadata corresponding to each of the sub-RAIDs; and in response to a disk in the RAID being replaced, rebuilding data in a corresponding sub-RAID based on the sub-RAID metadata.

2 . The method according to claim 1 , wherein the sub-RAID metadata includes a rebuilding state, the method comprising:
in response to a disk in the RAID being replaced, setting the rebuilding state to indicate that the disk needs to be rebuilt; rebuilding the disk according to the indication of the rebuilding state; and in response to completing the rebuilding of the disk, setting the rebuilding state to indicate that the disk does not need to be rebuilt.

3 . The method according to claim 2 , wherein each disk in the RAID corresponds to one bit of the rebuilding state, the method comprising:
setting a bit of the rebuilding state corresponding to the disk to indicate whether the disk needs to be rebuilt.

4 . The method according to claim 1 , wherein the sub-RAID metadata includes a rebuilding priority, the method comprising:
rebuilding data in the plurality of sub-RAIDs according to a priority order.

5 . The method according to claim 4 , further comprising:
generating RAID metadata corresponding to the RAID.

6 . The method according to claim 5 , wherein the RAID metadata includes a disk signature, the disk signature including a disk identifier uniquely identifying a disk and a RAID identifier uniquely identifying a RAID.

7 . The method according to claim 6 , further comprising:
in response to a disk in the RAID being replaced, determining whether the disk signature of a new disk exists in the RAID metadata; if it is determined that the disk signature of the new disk exists in the RAID metadata, directly connecting the new disk into the RAID; if it is determined that the disk signature of the new disk does not exist in the RAID metadata, updating the disk signature of the new disk with the RAID identifier of the RAID; and storing the updated disk signature in the RAID metadata.

8 . The method according to claim 5 , wherein the RAID metadata includes the number of sub-RAIDs that need to be rebuilt and a list of the sub-RAIDs that need to be rebuilt, the method further comprising:
in response to completing the rebuilding of data in a sub-RAID, updating the number of sub-RAIDs that need to be rebuilt and the list of the sub-RAIDs that need to be rebuilt.

9 . The method according to claim 1 , further comprising:
when the rebuilding of data in all the sub-RAIDs has not been completed and a further disk in the RAID becomes unavailable, determining whether the rebuilding state in the sub-RAID metadata indicates existence of a disk that needs to be rebuilt; if the rebuilding state in the sub-RAID metadata indicates that no disk needs to be rebuilt, recovering data in the sub-RAID; and if the rebuilding state in the sub-RAID metadata indicates existence of a disk that needs to be rebuilt, labeling data in the sub-RAID as being lost.

10 . The method according to claim 5 , wherein the sub-RAID metadata and the RAID metadata are stored in storage external to the RAID.

11 . An apparatus for a redundant array of independent disks, RAID, comprising:
a dividing unit configured to divide the RAID into a plurality of sub-RAIDs; a generating unit configured to generate sub-RAID metadata corresponding to each of the sub-RAIDs; and a rebuilding unit configured to, in response to a disk in the RAID being replaced, rebuilt data in a corresponding sub-RAID based on the sub-RAID metadata.

12 . The apparatus according to claim 11 , wherein the sub-RAID metadata includes a rebuilding state, the apparatus comprising:
a setting unit configured to, in response to a disk in the RAID being replaced, set the rebuilding state to indicate that the disk needs to be rebuilt; wherein the rebuilding unit is further configured to rebuild the disk that needs to be rebuilt according to the indication of the rebuilding state; and the setting unit is further configured to, in response to completing the rebuilding of the disk, set the rebuilding state to indicate that the disk does not need to be rebuilt.

13 . The apparatus according to claim 12 , wherein each disk in the RAID corresponds to one bit of the rebuilding state and the setting unit is further configured to:
set a bit of the rebuilding state corresponding to the disk to indicate whether the disk needs to be rebuilt.

14 . The apparatus according to claim 11 , wherein the sub-RAID metadata includes a rebuilding priority and the rebuilding unit is further configured to:
rebuild data in the plurality of sub-RAIDs according to a priority order.

15 . The apparatus according to claim 14 , wherein the generating unit is further configured to:
generate RAID metadata corresponding to the RAID.

16 . The apparatus according to claim 15 , wherein the RAID metadata includes a disk signature, the disk signature including a disk identifier uniquely identifying a disk and a RAID identifier uniquely identifying a RAID.

17 . The apparatus according to claim 16 , further comprising:
a determining unit configured to, in response to a disk in the RAID being replaced, determine whether the disk signature of a new disk exists in the RAID metadata; a connecting unit configured to, if it is determined that the disk signature of the new disk exists in the RAID metadata, directly connect the new disk into the RAID; and an updating unit configured to, if it is determined that the disk signature of the new disk does not exist in the RAID metadata, update the disk signature of the new disk with the RAID identifier of the RAID, and store the updated disk signature in the RAID metadata.

18 . The apparatus according to claim 15 , wherein the RAID metadata includes the number of sub-RAIDs that need to be rebuilt and a list of the sub-RAIDs that need to be rebuilt, and the updating unit is further configured to:
in response to completing the rebuilding of data in a sub-RAID, update the number of sub-RAIDs that need to be rebuilt and the list of the sub-RAIDs that need to the rebuilt.

19 . The apparatus according to claim 11 , wherein the determining unit is further configured to, when the rebuilding of data in all the sub-RAIDs has not been completed and a further disk in the RAID becomes unavailable, determine whether the rebuilding state in the sub-RAID metadata indicates existence of a disk that needs to be rebuilt; the apparatus further comprising:
a recovering unit configured to, if the rebuilding state in the sub-RAID metadata indicates that no disk needs to be rebuilt, recover data in the sub-RAID; and a labeling unit configured to, if the rebuilding state in the sub-RAID metadata indicates existence of a disk that needs to be rebuilt, label data in the sub-RAID as being lost.

20 . The apparatus according to claim 15 , wherein the sub-RAID metadata and the RAID metadata are stored in storage external to the RAID.

Assignees

• Emc Ip Holding Co Llc

Inventors

• Gao Hongpo
• Gao Jian
• Xiao Huibing
• Dong Jibing
• Han Geng

Classifications

• G06F11/1092 Primary

  Rebuilding, e.g. when physically replacing a failing disk · CPC title

• G06F2201/805

  Real-time · CPC title

• G06F11/1662 Primary

  the resynchronized component or unit being a persistent storage device (re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008) · CPC title

• G06F2201/82

  Solving problems relating to consistency · CPC title

• G06F11/2094

  Redundant storage or storage space (G06F11/2056 takes precedence) · CPC title