On-line restoration of redundancy information in a redundant array system

• US 5,235,601 A
• Filed: 12/21/1990
• Issued: 08/10/1993
• Est. Priority Date: 12/21/1990
• Status: Expired due to Term

• Current Assignee: EMC Corporation (Dell Technologies Inc.)
• Original Assignee: Array Technology Corp. (Dell Technologies Inc.)
• Inventors: Gordon, David, Brant, William A., Stallmo, David C.
• Primary Examiner(s): ATKINSON, CHARLES
• Application Number:
  US07/632,182
• Time in Patent Office: 963 Days
• Field of Search: 371/10.1, 371/10.2, 371/13, 371/40.1, 395/575
• US Class Current: 714/766
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1435 using file system or storage system metadata

Abstract

A method for on-line restoration of redundancy information in a redundant array storage system. The invention provides alternative methods of restoring valid data to a storage unit after a Write failure caused by a temporary storage unit fault. In the first preferred method, a valid redundancy block is generated for the corresponding data blocks on all storage units. Resubmitting the interrupted Write operation causes the old (and potentially corrupted) data block to be "subtracted" out of the re-computed redundancy block. The uncorrupted new data block is written over the old data block, and is "added" into the recomputed redundancy block to create a new, corrected redundancy block. The new, corrected redundancy block is written to the appropriate storage unit. In the second preferred method, a new redundancy block is generated from all valid data blocks and the new data block. The new redundancy block and the new data block are then written to the appropriate storage units. In both cases, the entire method is done on-line, with insignificant interruption of normal operation of the redundant array system, and without requiring added processing during normal operation.

Claims

1. In a redundant array of data storage units coupled to a controller, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, a method for restoration of a valid data block and at least one associated redundancy block to each data storage unit stripe after a potential corruption in either of such blocks caused by a temporary fault in a data storage unit during a data modification operation, comprising the steps of:

a. accessing all of the data blocks, including the potentially corrupted data block, in each stripe containing the potentially corrupted blocks;

b. computing at least one redundancy block from the accessed blocks;

c. saving the at least one computed redundancy block;

d. resubmitting the valid data block from the data modification operation to the redundant array of storage units for storage;

e. updating the at least one saved computed redundancy block;

f. storing the updated at least one redundancy block and the valid data block in the stripe.

2. The method of claim 1, wherein at least one redundancy block in each stripe contains parity information, and the step of computing the parity redundancy block comprises exclusively-OR'"'"'ing the accessed blocks.

3. The method of claim 1, wherein at least one saved computed redundancy block contains parity information, and the step of updating that parity redundancy block comprises the steps of:

a. accessing the potentially corrupted data block; and

b. exclusively-OR'"'"'ing the accessed data block with the saved parity redundancy block and the resubmitted valid data block to generate a new parity redundancy block.

4. The method of claim 1, further including the step of performing the steps of claim 1 as a task concurrently with other input/output tasks.

5. The method of claim 4, further including the step of locking each block being read or modified during restoration.

6. The method of claim 1, further including the step of storing each data modification operation submitted to the redundant array in a non-volatile storage device until the data modification operation is completed.

7. In a redundant array of data storage units coupled to a controller, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, a method for on-line restoration of a valid data block and at least one associated redundancy block to each data storage unit stripe after a potential corruption in either of such blocks caused by a temporary fault in a data storage unit during a data modification operation, comprising the steps of:

a. resubmitting the valid data block from the data modification operation to the redundant array of storage units for storage;

b. accessing all of the uncorrupted data blocks in the stripe containing the potentially corrupted blocks;

c. computing at least one redundancy block from the accessed blocks and the resubmitted valid data block;

d. storing the computed at least one redundancy block and the valid data block in the stripe;

e. performing restoration as a task concurrently with other input/output tasks.

8. The method of claim 7, wherein at least one redundancy block in each stripe contains parity information, and the step of computing the parity redundancy block comprises exclusively-OR'"'"'ing the accessed blocks with the resubmitted valid data block.

9. The method of claim 7, further including the step of locking each block being read or modified during restoration.

10. The method of claim 7, further including the step of storing each data modification operation submitted to the redundant array in a non-volatile storage device until the data modification operation is completed.

11. In a redundant array of data storage units coupled to a controller, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, a method for restoration of at least one valid redundancy block to each data storage unit strip after a potential corruption in a data block and/or in such redundancy block in such stripe caused by a temporary fault in a data storage unit during a data modification operation, comprising the steps of:

a. accessing all of the data blocks, including the potentially corrupted data block, in each stripe containing the potentially corrupted blocks;

b. computing at least one redundancy block from the accessed blocks;

c. saving the at least one computed redundancy block in the stripe.

12. The method of claim 11, wherein at least one redundancy block in each stripe contains parity information, and the step of computing the parity redundancy block comprises exclusively-OR'"'"'ing the accessed blocks.

13. The method of claim 11, further including the step of performing the steps of claim 12 as a task concurrently with other input/output tasks.

14. The method of claim 13, further including the step of locking each block being read or modified during restoration.

15. The method of claim 11, further including the step of storing each data modification operation submitted to the redundant array in a non-volatile storage device until the data modification operation is completed.

16. A control system for use in conjunction with a redundant array of data storage units, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, the control system for restoration of a valid data block and at least one associated redundancy block to each data storage unit stripe after a potential corruption in either of such blocks caused by a temporary fault in a data storage unit during a data modification operation, the control system including:

a. coupling means for coupling the control system to the array of storage units;

b. restoration means, coupled to the coupling means, for;

(1) accessing all of the data blocks, including the potentially corrupted data block, in each stripe containing the potentially corrupted blocks;

(2) computing at least one redundancy block from the accessed blocks;

(3) saving the at least one computed redundancy block;

(4) resubmitting the valid data block from the data modification operation to the redundant array of storage units for storage;

(5) updating the at least one saved computed redundancy block;

(6) storing the updated at least one redundancy block and the valid data block in the stripe.

17. The control system of claim 16, wherein at least one redundancy block in each stripe contains partity information, and the parity redundancy block is computed by exclusively-OR'"'"'ing the accessed blocks.

18. The control system of claim 16, wherein at least one saved computed redundancy block contains parity information, and the control system further includes means for updating that parity redundancy block by:

a. accessing the potentially corrupted data block; and

b. exclusively-OR'"'"'ing the accessed data block with the saved parity redundancy block and the resubmitted valid data block to generate a new parity redundancy block.

19. The control system of claim 16, wherein the control system includes means for performing restoration as a task concurrently with other input/output tasks.

20. The control system of claim 19, wherein the control system further includes means for locking each block being read or modified during restoration.

21. The control system of claim 16, further including a non-volatile storage device, coupled to the control system, for storing each data modification operation submitted to the redundant array until the data modification operation is completed.

22. A control system for use in conjunction with a redundant array of data storage units, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, the control system for on-line restoration of a valid data block and at least one associated redundancy block to each data storage unit stripe after a potential corruption in either of such blocks caused by a temporary fault in a data storage unit during a data modification operation, the control system including:

a. coupling means for coupling the control system to the array of storage units;

b. restoration means, coupled to the coupling means, for;

(1) resubmitting the valid data block from the data modification operation to the redundant array of storage units for storage;

(2) accessing all of the uncorrupted data blocks in the stripe containing the potentially corrupted blocks;

(3) computing at least one redundancy block from the accessed blocks and the resubmitted valid data block;

(4) storing the computed at least one redundancy block and the valid data block in the stripe;

(5) performing restoration as a task concurrently with other input/output tasks.

23. The control means of claim 22, wherein at least one redundancy block in each stripe contains parity information, and the parity redundancy block is computed by exclusively-OR'"'"'ing the accessed blocks with the resubmitted valid data block.

24. The control system of claim 22, wherein the control system further includes means for locking each block being read or modified during restoration.

25. The control system of claim 22, further including a non-volatile storage device, coupled to the control system, for storing each data modification operation submitted to the redundant array until the data modification operation is completed.

26. A control system for use in conjunction with a redundant array of data storage units, the data storage units having at least one stripe containing a plurality of data blocks and at least one associated redundancy block, the control system for restoration of at least one valid redundancy block to each data storage unit stripe after a potential corruption in a data block and/or in such redundancy block in such stripe caused by a temporary fault in a data storage unit during a data modification operation, the control system including:

a. coupling means for coupling the control system to the array of storage units;

b. restoration means, coupled to the coupling means, for;

(1) accessing all of the data blocks, including the potentially corrupted data block, in each stripe containing the potentially corrupted blocks;

(2) computing at least one redundancy block from the accessed blocks;

(3) saving the at least one computed redundancy block in the stripe.

27. The control system of claim 26, wherein at least one redundancy block in each stripe contains parity information, and the parity redundancy block is computed by exclusively OR'"'"'ing the accessed blocks.

28. The control system of claim 26, wherein the control system includes means for performing restoration as a task concurrently with other input/output tasks.

29. The control system of claim 28, wherein the control system further includes means for locking each block being read or modified during restoration.

30. The control system of claim 30, further including a non-volatile storage device, coupled to the control system, for storing each data modification operation submitted to the redundant array until the data modification operation is completed.