N-WAY PARITY TECHNIQUE FOR ENABLING RECOVERY FROM UP TO N STORAGE DEVICE FAILURES

• US 20110010599A1
• Filed: 09/25/2009
• Published: 01/13/2011
• Est. Priority Date: 12/28/2001
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Goel, Atul, Corbett, Peter F.
• Granted Patent: US 8,402,346 B2
• Time in Patent Office: Days
• US Class Current: 714/752
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems

Abstract

An n-way parity protection technique enables recovery of up to n storage device (e.g., disk) failures in a parity group of a storage array encoded to protect against n-way disk failures. The storage array is created by first configuring the array with m data disks, where m=p−1 and p is a prime number and a row parity disk. n−1 diagonal parity disks are then added to the array. Each diagonal parity set (i.e., diagonal) is associated with a slope that defines the data and row parity blocks of the array that are included in the diagonal. All diagonals having a common slope within a parity group are organized as a diagonal parity class. For each diagonal parity class, a diagonal parity storage disk is provided to store the diagonal parity.

Claims

1. A method for enabling recovery from up to n concurrent failures of storage devices in a storage array, comprising:

providing the array with a predetermined number of storage devices, including a plurality of first devices configured to store data and row parity, wherein the predetermined number of storage devices m is less than or equal to p−

1 and wherein p is a prime number;

providing the array with at least three second devices configured to store at least three diagonal parity classes;

dividing each device into blocks;

organizing the blocks into stripes that contain blocks in each device;

computing a row parity for each row of data;

assigning all blocks from the devices storing data and row parity to diagonals; and

for each diagonal parity class, computing diagonal parity along all diagonals having a common slope and storing the computed diagonal parity on one of the second devices associated with the diagonal parity class.

2. The method of claim 1 wherein row parity blocks in a stripe are all stored on a single device.

3. The method of claim 1 wherein the first devices include a plurality of data devices, and wherein the data devices are not all present and absent data devices are treated as containing all zeroes for calculating parity.

4. The method of claim 1 wherein locations of parity blocks shift from device to device within different stripes.

5. The method of claim 1 wherein the stripes are a power of two number of bits in size.

6. The method of claim 1 wherein the plurality of second devices comprises up to n−

1 devices and wherein n≦

p.

7. The method of claim 1 wherein the first and second plurality of storage devices comprise disk drives.

8. The method of claim 1 further comprising, in response to detecting failure of one or more storage devices, performing a reconstruction technique on the non-failed storage devices.

9. The method of claim 1 further comprising:

adding an additional second device to store an additional diagonal parity class wherein only the diagonal parity along all diagonals having a common slope with the additional diagonal parity class is computed; and

storing the computed diagonal parity on the additional second device associated with the additional diagonal parity class.

10. The method of claim 1 wherein the slope comprises a ratio of a number of storage devices over and up between two adjacent storage devices in a diagonal.

11. The method of claim 1 wherein the common slope comprises slopes being equal in value.

12. The method of claim 1 wherein the storage devices comprise flash devices.

13. A system to enable recovery from up to n concurrent failures of storage devices in a storage array, comprising:

the storage array configured with a predetermined number of storage devices, including a plurality of first devices configured to store data and row parity, and a plurality of second devices configured to store diagonal parity, wherein the predetermined number of storage devices m is p−

1 and wherein p is a prime number, wherein each device is divided into blocks and the blocks are organized into stripes that contain a same number of blocks in each device;

the storage array further configured with a at least three second devices configured to store diagonal parity classes, each diagonal parity class defined by a slope along the data and row parity; and

a storage operating system including a device storage module configured to compute and store diagonal parity for all of the diagonal parity classes.

14. The system of claim 13, wherein row parity blocks in a stripe are all stored on a single device.

15. The system of claim 13, wherein the device storage module layer is a RAID system and wherein the storage devices are disks.

16. The system of claim 13, wherein the devices are one of video tape, magnetic tape, optical, DVD, bubble memory, magnetic disk, electronic random access memory and micro-electro mechanical storage devices.

17. The system of claim 13, wherein the first devices include a plurality of data devices, and wherein the data devices are not all present and absent data devices are treated as containing all zeroes for calculating parity.

18. The system of claim 13, wherein row parity blocks are rotated through a plurality of devices in a stripe.

19. The system of claim 13, wherein the device storing row parity, diagonal parity, or data varies from stripe to stripe.

20. The system of claim 13 wherein a number of distinct slopes utilized is at most p−

1.

21. The system of claim 13 wherein the device storage module is further configured to reconstruct data from a plurality of failures of the first or second plurality of devices.

22. The system of claim 13 wherein the storage devices are flash devices.

23. A method for enabling recovery from up to n concurrent failures of storage devices in a storage array, comprising:

associating a set of slopes with the storage array, each slope associated with a diagonal parity class of a set of data and row parity devices within the storage array, wherein no two slopes are equal module a prime number associated with the storage array;

determining, by a storage module of a storage operating system executing on a storage system, row parity for each row of data within the storage array and storing the determined row parity on the set of row parity devices; and

determining, for each diagonal parity class, a set of diagonal parity that spans the data and row parity devices and storing the determined diagonal parity on a diagonal parity device associated with the diagonal parity class within the storage array.

24. The method of claim 23 wherein the storage devices comprise disk drives.

25. The method of claim 23 wherein the set of data and row parity devices comprise p device, wherein p is a prime number and wherein the set of diagonal parity devices comprises n−

1 devices.