Sparse raid rebuild based on storage extent allocation

Patent metadata

Field	Value
Publication number	US-9804939-B1
Application number	US-201514870190-A
Country	US
Kind code	B1
Filing date	Sep 30, 2015
Priority date	Sep 30, 2015
Publication date	Oct 31, 2017
Grant date	Oct 31, 2017

Abstract

A technique for repairing a RAID (Redundant Array of Independent Disks) group in a data storage system following a failure of a disk drive in the RAID group includes identifying regions of the failed disk drive that supported unallocated storage extents and writing predetermined data to corresponding regions of the spare disk drive. Any reading of other disk drives in the RAID group may thus be limited only to those regions of the failed disk drive that supported allocated storage extents.

First claim

What is claimed is:

1. A method of repairing a RAID (Redundant Array of Independent Disks) group in a data storage system following a failure of a disk drive in the RAID group, the method comprising:
providing a spare disk drive to replace the disk drive in the RAID group that failed;
accessing metadata in the data storage system, the metadata indicating an allocation status of multiple storage extents formed from the RAID group, the storage extents formed from the RAID group including a first set of storage extents that the metadata indicates is not allocated and a second set of storage extents that the metadata indicates is allocated;
for each of the first set of storage extents, writing predetermined data to the spare disk drive at regions that will support that storage extent; and
for each of the second set of storage extents, (i) reading portions of data from regions that support that storage extent on a set of other disk drives of the RAID group, (ii) providing repair data for that storage extent based on the portions of data read for that storage extent, and (iii) writing the repair data for that storage extent to the spare disk drive,
the method thereby avoiding any need to read portions of data from other disk drives for storage extents that are not allocated.

2. The method of claim 1 ,
wherein the storage extents are uniformly-sized extents each configured to store multiple blocks of data,
wherein the second set of storage extents have been allocated by the data storage system for providing storage space for one or more data objects hosted by the data storage system.

3. The method of claim 2 , wherein each of the first set of storage extents stores initial data, and wherein the predetermined data, written to the spare disk, is based on the initial data.

4. The method of claim 3 ,
wherein the RAID group is arranged in a RAID 1 configuration, and
wherein the predetermined data written for each of the first set of storage extents is the same as the initial data.

5. The method of claim 3 ,
wherein the initial data is all zeros,
wherein each of the first set of storage extents includes an integer number of stripes, each stripe having multiple data elements and a parity element, and
wherein the predetermined data written for each of the first set of storage extents is all zeros for each of the data elements in each stripe and is all zeros for the parity element in each stripe.

6. The method of claim 3 ,
wherein the initial data is all zeros,
wherein each of the first set of storage extents includes an integer number of stripes, each stripe having multiple data elements, a first parity element, and a second parity element, and
wherein the predetermined data written for each of the first set of storage extents is all zeros for each of the data elements in each stripe, is a second predetermined value for each first parity element in each stripe, and is a third predetermined value for each second parity element in each stripe.

7. The method of claim 3 , further comprising mapping, by the data storage system, the storage extents to respective disk drive locations where data of the storage extents are stored.

8. The method of claim 7 , further comprising:
receiving a request to free a previously allocated storage extent; and
in response to receiving the request, (i) writing the initial data to the previously allocated storage extent and (ii) updating the metadata to indicate that the previously allocated storage extent is not allocated.

9. The method of claim 8 , wherein, when writing the predetermined data to the spare disk drive for each of the first set of storage extents, the method further comprises executing a single write instruction to write the predetermined data across multiple ones of the first set of storage extents.

10. A data storage system comprising control circuitry that includes a set of processing units coupled to memory, the control circuitry constructed and arranged to:
provide a spare disk drive to replace the disk drive in the RAID group that failed;
access metadata in the data storage system, the metadata indicating an allocation status of multiple storage extents formed from the RAID group, the storage extents formed from the RAID group including a first set of storage extents that the metadata indicates is not allocated and a second set of storage extents that the metadata indicates is allocated; and
for each of the first set of storage extents, write predetermined data to the spare disk drive at regions that will support that storage extent;
for each of the second set of storage extents, (i) read portions of data from regions that support that storage extent on a set of other disk drives of the RAID group, (ii) provide repair data for that storage extent based on the portions of data read for that storage extent, and (iii) write the repair data for that storage extent to the spare disk drive.

11. The data storage system of claim 10 ,
wherein the storage extents are uniformly-sized extents each configured to store multiple blocks of data,
wherein the second set of storage extents have been allocated by the data storage system for providing storage space for one or more data objects hosted by the data storage system.

12. A computer program product including a set of non-transitory, computer-readable media having instructions which, when executed by control circuitry of a data storage system, cause the control circuitry to perform a method for repairing a RAID (Redundant Array of Independent Disks) group in a data storage system following a failure of a disk drive in the RAID group, the method comprising:
providing a spare disk drive to replace the disk drive in the RAID group that failed;
accessing metadata in the data storage system, the metadata indicating an allocation status of multiple storage extents formed from the RAID group, the storage extents formed from the RAID group including a first set of storage extents that the metadata indicates is not allocated and a second set of storage extents that the metadata indicates is allocated;
for each of the first set of storage extents, writing predetermined data to the spare disk drive at regions that will support that storage extent; and
for each of the second set of storage extents, (i) reading portions of data from regions that support that storage extent on a set of other disk drives of the RAID group, (ii) providing repair data for that storage extent based on the portions of data read for that storage extent, and (iii) writing the repair data for that storage extent to the spare disk drive,
the method thereby avoiding any need to read portions of data from other disk drives for storage extents that are not allocated.

13. The computer program product of claim 12 ,
wherein the storage extents are uniformly-sized extents each configured to store multiple blocks of data,
wherein the second set of storage extents have been allocated by the data storage system for providing storage space for one or more data objects hosted by the data storage system.

14. The computer program product of claim 13 , wherein each of the first set of storage extents stores initial data, and wherein the predetermined data, written to the spare disk, is based on the initial data.

15. The computer program product of claim 14 ,
wherein the RAID group is arranged in a RAID 1 configuration, and
wherein the predetermined data written for each of the first set of storage extents is the same as the initial data.

16. The computer program product of claim 14 ,
wherein the initial data is all zeros,
wherein each

Assignees

• Emc Corp
• Emc Ip Holding Co Llc

Inventors

• Bono Jean-Pierre
• Armangau Philippe

Classifications

• G06F11/2094 Primary

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

• G06F11/1088

  Reconstruction on already foreseen single or plurality of spare disks · CPC title

• G06F3/0644

  Management of space entities, e.g. partitions, extents, pools · CPC title

• G06F3/0689

  Disk arrays, e.g. RAID, JBOD · CPC title

• G06F3/0619 Primary

  in relation to data integrity, e.g. data losses, bit errors · CPC title