Error checking addressable blocks in storage

• US 8,370,715 B2
• Filed: 04/12/2007
• Issued: 02/05/2013
• Est. Priority Date: 04/12/2007
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Hafner, James Lee, Kahler, David Ray, Kubo, Robert Akira, Mannenbach, David Frank, Nielsen, Karl Allen, O'Connor, James A., Surugucchi, Krishnakumar Rao, Stelmach, Richard B.
• Primary Examiner(s): Torres, Joseph D
• Application Number:
  US11/734,786
• Publication Number:
  US 20080256420A1
• Time in Patent Office: 2,126 Days
• Field of Search: 714/769, 714/770
• US Class Current: 714/769
• CPC Class Codes: G06F 11/1004 to protect a block of data words, e.g. CRC or checksum G06F11/1076 takes precedence; security arrangements for protecting computers or computer systems against unauthorized activity G06F21/00
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems

Abstract

Provided are a method, system, and article of manufacture for error checking addressable blocks in storage. Addressable blocks of data are stored in a storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block. A write request is received to modify data in one of the addressable blocks. The write and updating the checksum are performed in the stripe having the modified addressable block. An indication is made to perform an error checking operation on the stripe for the modified addressable block in response to the write request, wherein the error checking operation reads the data blocks and the checksum in the stripe to determine if the checksum data is accurate. An error handling operation is initiated in response to determining that the checksum data is not accurate.

Claims

1. An article of manufacture comprising a non-transitory computer readable storage medium having code executed to cause operations with respect to a storage device and a memory and to cause operations comprising:

storing addressable blocks of data in the storage device in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block stored in the stripe having the addressable block;

receiving a write request to modify data in one of the addressable blocks in the storage device;

performing the write and updating the at least one checksum block in the stripe having the modified addressable block, wherein the modified data and the updated at least one checksum block resulting from the write are destaged to the storage device;

indicating in a recently modified list in the memory, the addressable block modified in response to the write request;

performing an error checking operation on the stripes of the addressable blocks by performing;

determining whether the stripe having the addressable block is indicated in the recently modified list;

using the at least one checksum block for the stripe having the addressable block to correct the error in response to determining that the addressable block of the stripe is indicated in the recently modified list; and

indicating the stripe having the addressable block as having an uncorrectable error in response to determining that the addressable block of the stripe is not indicated in the recently modified list.

2. The article of manufacture of claim 1, wherein the operations further comprise:

recalculating the checksum data in the stripe as part of the error checking operation or a write during which the at least one checksum block is recalculated using all the data blocks in the stripe; and

removing the addressable block from the recently modified list in response to recalculating the at least one checksum block using all the data blocks in the stripe of the addressable block.

3. The article of manufacture of claim 1, wherein the operations further comprise:

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data.

4. The article of manufacture of claim 1, wherein the data and checksum blocks in one stripe are written to different storage devices according to a Redundant Array of Independent Disk (RAID) algorithm.

5. The article of manufacture of claim 1, wherein the error checking operation comprises a first error checking operation, wherein the operations further comprise:

performing a second error checking operation for addressable blocks indicated in the recently modified list;

determining whether an error is detected for any of the stripes for addressable blocks indicated in the modified list during the second error checking operation;

initiating an error handling operation with respect to the stripes for which the error is detected during the second error checking operation; and

removing the addressable blocks for the stripes from the modified list for which the error checking operation is performed.

6. The article of manufacture of claim 1, wherein the operations performed for the stripes further comprise:

determining whether the checked stripe has a detectable error as a result of the error checking operation; and

initiating an error handling operation in response to determining that the stripe has the detectable error.

7. The article of manufacture of claim 6, wherein the operation of determining that the stripe having the addressable block is indicated in the recently modified list is performed in response to determining that the stripe does not have the detectable error.

8. An article of manufacture comprising a non-transitory computer readable storage medium having code executed to cause operations with respect to a storage device and a memory and to cause operations comprising:

storing addressable blocks of data in the storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block;

receiving a write request to modify data in one of the addressable blocks;

performing the write and updating the checksum data in the stripe having the modified addressable block;

indicating to perform an error checking operation on the stripe for the modified addressable block in response to the write request by indicating the modified addressable block in a modified list in the memory, wherein the error checking operation processes the modified list to read the data blocks and the checksum data in the stripe to determine if the checksum data for the addressable blocks in the modified list is accurate;

initiating an error handling operation in response to determining that the checksum data is not accurate;

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data; and

alternating between processing m stripes as part of the background operation and n stripes as part of the error checking operation processing modified addressable blocks indicated in the modified list, wherein m is greater than n.

9. A system in communication with a storage device, comprising:

a memory;

storage manager code enabled to cause operations, the operations comprising;

storing addressable blocks of data in the storage device in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block stored in the stripe having the addressable block;

receiving a write request to modify data in one of the addressable blocks in the storage device;

performing the write and updating the at least one checksum block in the stripe having the modified addressable block, wherein the modified data and the updated at least one checksum block resulting from the write are destaged to the storage device;

indicating in a recently modified list in the memory the addressable block modified in response to the write request;

performing an error checking operation on the stripes of the addressable blocks by performing;

determining whether the stripe having the addressable block is indicated in the recently modified list;

using the at least one checksum block for the stripe having the addressable block to correct the error in response to determining that the addressable block of the stripe is indicated in the recently modified list; and

indicating the stripe having the addressable block as having an uncorrectable error in response to determining that the addressable block of the stripe is not indicated in the recently modified list.

10. The system of claim 9, wherein the operations further comprise:

recalculating the checksum data in the stripe as part of the error checking operation or a write during which the at least one checksum block is recalculated using all the data blocks in the stripe; and

removing the addressable block from the recently modified list in response to recalculating the at least one checksum block using all the data blocks in the stripe of the addressable block.

11. The system of claim 9, wherein the operations further comprise:

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data.

12. The system of claim 9, wherein the error checking operation comprises a first error checking operation, wherein the operations further comprise:

performing a second error checking operation for addressable blocks indicated in the recently modified list;

determining whether an error is detected for any of the stripes for addressable blocks indicated in the modified list during the second error checking operation;

initiating an error handling operation with respect to the stripes for which the error is detected during the second error checking operation; and

removing the addressable blocks for the stripes from the modified list for which the error checking operation is performed.

13. The system of manufacture of claim 9, wherein the operations performed for the stripes further comprise:

determining whether the checked stripe has a detectable error as a result of the error checking operation; and

initiating an error handling operation in response to determining that the stripe has the detectable error.

14. The system of claim 13, wherein the operation of determining that the stripe having the addressable block is indicated in the recently modified list is performed in response to determining that the stripe does not have the detectable error.

15. A system in communication with a storage, comprising:

a memory;

storage manager code enabled to cause operations, the operations comprising;

storing addressable blocks of data in the storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block;

receiving a write request to modify data in one of the addressable blocks;

performing the write and updating the checksum data in the stripe having the modified addressable block;

indicating to perform an error checking operation on the stripe for the modified addressable block in response to the write request by indicating the modified addressable block in a modified list in the memory, wherein the error checking operation processes the modified list to read the data blocks and the checksum data in the stripe to determine if the checksum data for addressable blocks in the modified list is accurate;

initiating an error handling operation in response to determining that the checksum data is not accurate;

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data; and

alternating between processing m stripes as part of the background operation and n stripes as part of the error checking operation processing modified addressable blocks indicated in the modified list, wherein m is greater than n.

16. A method, comprising:

storing addressable blocks of data in a storage device in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block stored in the stripe having the addressable block;

receiving a write request to modify data in one of the addressable blocks in the storage device;

performing the write and updating the at least one checksum block in the stripe having the modified addressable block, wherein the modified data and the updated at least one checksum block resulting from the write are destaged to the storage device;

indicating in a recently modified list in the memory the addressable block recently modified in response to the write request;

performing an error checking operation on the stripes of the addressable blocks by performing;

determining whether the stripe having the addressable block is indicated in the recently modified list;

using the at least one checksum block for the stripe having the addressable block to correct the error in response to determining that the addressable block of the stripe is indicated in the recently modified list; and

indicating the stripe having the addressable block as having an uncorrectable error in response to determining that the addressable block of the stripe is not indicated in the recently modified list.

17. The method of claim 16, further comprising:

recalculating the checksum data in the stripe as part of the error checking operation or a write during which the at least one checksum block is recalculated using all the data blocks in the stripe; and

removing the addressable block from the recently modified list in response to recalculating the at least one checksum block using all the data blocks in the stripe of the addressable block.

18. The method of claim 16, further comprising

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data.

19. The method of claim 16, wherein the error checking operation comprises a first error checking operation, further comprising:

performing a second error checking operation for addressable blocks indicated in the recently modified list;

determining whether an error is detected for any of the stripes for addressable blocks indicated in the modified list during the second error checking operation;

initiating an error handling operation with respect to the stripes for which the error is detected during the second error checking operation; and

removing the addressable blocks for the stripes from the modified list for which the error checking operation is performed.

20. The method of claim 16, wherein the operations performed for the stripes further comprise:

determining whether the checked stripe has a detectable error as a result of the error checking operation; and

initiating an error handling operation in response to determining that the stripe has the detectable error.

21. The method of claim 20, wherein the operation of determining that the stripe having the addressable block is indicated in the recently modified list is performed in response to determining that the stripe does not have the detectable error.

22. A method, comprising:

storing addressable blocks of data in a storage in stripes, wherein each stripe includes a plurality of data blocks for one of the addressable blocks and at least one checksum block including checksum data derived from the data blocks for the addressable block;

receiving a write request to modify data in one of the addressable blocks;

performing the write and updating the checksum data in the stripe having the modified addressable block;

indicating to perform an error checking operation on the stripe for the modified addressable block in response to the write request by indicating the modified addressable block in a modified list in the memory, wherein the error checking operation reads the data blocks and the checksum data in the stripe to determine if the checksum data for addressable blocks in the modified list is accurate;

initiating an error handling operation in response to determining that the checksum data is not accurate;

initiating a background operation to read the data blocks and the at least one checksum block in each stripe for a plurality of addressable blocks to determine if the checksum data is accurate, wherein the plurality of addressable blocks on which to perform the background operation includes modified and unmodified data; and

alternating between processing m stripes as part of the background operation and n stripes as part of the error checking operation processing modified addressable blocks indicated in the modified list, wherein m is greater than n.