Self-healing log-structured RAID

• US 20030120863A1
• Filed: 12/09/2002
• Published: 06/26/2003
• Est. Priority Date: 12/26/2001
• Status: Active Grant

• Current Assignee: BOON Storage Technologies Incorporated
• Original Assignee: BOON Storage Technologies Incorporated
• Inventors: Yeo, Boon-Lock, Lee, Edward K.
• Granted Patent: US 7,055,058 B2
• Time in Patent Office: Days
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1084 Degraded mode, e.g. caused by single or multiple storage removals or disk failures
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system
• CPC Class Codes: G06F 2211/1004 Adaptive RAID, i.e. RAID system adapts to changing circumstances, e.g. RAID1 becomes RAID5 as disks fill up
• CPC Class Codes: G06F 2211/1038 LFS, i.e. Log Structured File System used in RAID systems with parity

Abstract

This invention describes a high-performance, log-structured implementation of a RAID subsystem that can be efficiently implemented in software. The methods described in this invention are extraordinarily well suited for software-based RAID 5 systems. In one embodiment, the disk storage subsystems receiving a request to write data. Thereafter, the write operation is appended to a first log to identify a storage location of data written to the disk subsystem, wherein the first log consists of a sequence of segments, the segments consist of a sequence of variable length stripes, a length of a stripe varies based on a quantity of disks addressed by a stripe. In response to adding a disk to the disk storage subsystem, increasing a length of the stripes and increasing a number of disks available for striping data. Alternatively, in response to a disk failing within the disk storage subsystem or a disk being removed from the disk storage system, narrowing a length of the stripes.

Claims

1. ) A method of storing data on a disk storage subsystem comprising:

Receiving a request to write data;

Appending each write operation to a first log to identify a storage location of data written to the disk subsystem, wherein the first log consists of a sequence of segments, the segments consist of a sequence of variable length stripes, a length of a stripe varies based on a quantity of disks addressed by a stripe; and

in response to adding a disk to the disk storage subsystem, increasing a length of the stripes and increasing a number of disks available for striping data.

2. ) The method of claim 1, wherein in response to a disk being removed, decreasing a length of the stripe.

3. ) The method of claim 2, wherein in further response to a disk being removed, treating the removed disk as if the removed disked had failed and performing disk failure recovery, and continuing to service read request to the disk until the disk is physically removed.

4. ) The method of claim 1, further including in response to a disk failure, deferring rebuild process until a predetermined time to tolerate multiple disk failures.

5. ) The method of claim 1, further including, in response to a disk failure, reading stripes that span a failed disk and appending the data within the stripes of the failed disk to the end of the first log.

6. ) A method of storing data on a disk storage subsystem comprising:

Receiving a request to write data;

Appending each write operation to a first log to identify a storage location of data written to the disk subsystem, wherein the log consists of a sequence of segments, the segments consist of a sequence of variable length stripes, a length of a stripe varies based on a quantity of disks addressed by a stripe; and

in response to a disk failing within the disk storage subsystem or a disk being removed from the disk storage system, narrowing a length of the stripes.

7. ) The method of claim 6, wherein in further response to a disk being removed, treating the removed disk as if the removed disked had failed and performing disk failure recovery, and continuing to service read request to the disk until the disk is physically removed.

8. ) The method of claim 6, further including in response to a disk failure, deferring rebuild process until a predetermined time to tolerate multiple disk failures.

9. ) The method of claim 6, further including, in response to a disk failure, reading stripes that span a failed disk and appending the data within the stripes of the failed disk to the end of the first log.

10. ) The method of claim 6, further including in response to adding a disk to the disk storage subsystem, increasing a length of the stripes and increasing a number of disks available for striping data.

21. ) The method of claim 6, wherein in response to a disk failure, not writing additional data to the disk and performing disk failure recovery;

and

In response to the failed disk recovering, the disk is then included in new writes.

11. ) A machine readable-medium having stored there on a set of instructions, which when executed, perform a method of storing data on a disk storage subsystem comprising:

Receiving a request to write data;

Appending each write operation to a first log to identify a storage location of data written to the disk subsystem, wherein the first log consists of a sequence of segments, the segments consist of a sequence of variable length stripes, a length of a stripe varies based on a quantity of disks addressed by a stripe; and

in response to adding a disk to the disk storage subsystem, increasing a length of the stripes and increasing a number of disks available for striping data.

12. ) The machine readable-medium of claim 11, wherein in response to a disk being removed, decreasing a length of the stripe.

13. ) The machine readable-medium of claim 12, wherein in further response to a disk being removed, treating the removed disk as if the removed disked had failed and performing disk failure recovery, and continuing to service read request to the disk until the disk is physically removed.

14. ) The machine readable-medium of claim 11, further including in response to a disk failure, deferring rebuild process until a predetermined time to tolerate multiple disk failures.

15. ) The machine readable-medium of claim 11, further including, in response to a disk failure, reading-stripes that span a failed disk and appending the data within the stripes of the failed disk to the end of the first log.

16. ) A machine readable-medium having stored there on a set of instructions, which when executed, perform a method of storing data on a disk storage subsystem comprising:

Receiving a request to write data;

Appending each write operation to a first log to identify a storage location of data written to the disk subsystem, wherein the log consists of a sequence of segments, the segments consist of a sequence of variable length stripes, a length of a stripe varies based on a quantity of disks addressed by a stripe; and

in response to a disk failing within the disk storage subsystem or a disk being removed from the disk storage system, narrowing a length of the stripes.

17. ) The machine readable-medium of claim 16, wherein in further response to a disk being removed, treating the removed disk as if the removed disked had failed and performing disk failure recovery, and continuing to service read request to the disk until the disk is physically removed.

18. ) The machine readable-medium of claim 16, further including in response to a disk failure, deferring rebuild process until a predetermined time to tolerate multiple disk failures.

19. ) The machine readable-medium of claim 16, further including, in response to a disk failure, reading stripes that span a failed disk and appending the data within the stripes of the failed disk to the end of the first log.

20. ) The machine readable-medium of claim 16, further including in response to adding a disk to the disk storage subsystem, increasing a length of the stripes and increasing a number of disks available for striping data.

22. ) The machine readable-medium of claim 16, wherein in response to a disk failure, not writing additional data to the disk and performing disk failure recovery;

and

In response to the failed disk recovering, the disk is then included in new writes.