Apparatus, system, and method for managing data storage

• US 8,412,978 B2
• Filed: 05/08/2012
• Issued: 04/02/2013
• Est. Priority Date: 05/16/2008
• Status: Expired due to Fees

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: Fusion-IO Incorporated (Western Digital Corporation)
• Inventors: Flynn, David, Thatcher, Jonathan, Aune, Joshua, Fillingim, Jeremy, Inskeep, Bill, Strasser, John, Vigor, Kevin
• Primary Examiner(s): BONZO, BRYCE P
• Application Number:
  US13/466,845
• Publication Number:
  US 20120221924A1
• Time in Patent Office: 329 Days
• Field of Search: 714/6.1
• US Class Current: 714/6.1
• CPC Class Codes: G06F 11/1008 in individual solid state devices G06F11/1004 takes precedence
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 2211/109 Sector level checksum or ECC, i.e. sector or stripe level checksum or ECC in addition to the RAID parity calculation
• CPC Class Codes: G11C 29/765 in solid state disks

Abstract

An apparatus, system, and method are disclosed for managing data storage. The method includes determining that an error correcting code (ECC) block comprises uncorrectable errors. The ECC block is stored across a plurality of memory devices. The method includes iteratively substituting replacement data, within data of the ECC block, for individual memory devices of the plurality of memory devices to form substitute ECC blocks until one of the substitute ECC blocks is correctable using the error correcting code for the ECC block. The method includes providing corrected data from the correctable one of the substitute ECC blocks.

Claims

1. A method for managing data storage, the method comprising:

determining that an error correcting code (ECC) block comprises uncorrectable errors, the ECC block stored across a plurality of memory devices;

iteratively substituting replacement data, within data of the ECC block, for individual memory devices of the plurality of memory devices to form substitute ECC blocks until one of the substitute ECC blocks is correctable using the error correcting code for the ECC block; and

providing corrected data from the correctable one of the substitute ECC blocks.

2. The method of claim 1, wherein the plurality of memory devices comprise at least a portion of an array of memory devices and at least one of the memory devices of the array stores parity information derived from data stored on each memory device of the plurality of memory devices storing the ECC chunk.

3. The method of claim 2, wherein the replacement data is derived from the parity information for individual memory devices of the array.

4. The method of claim 1, wherein the ECC chunk comprises both data and the error correcting code derived from the data.

5. The method of claim 1, further comprising retiring one or more storage regions on the failed memory device associated with the substitute ECC chunk, the one or more storage regions comprising one or more errors.

6. The method of claim 1, further comprising determining that a memory device that stored data of the ECC block is a failed memory device and generating data to replace data read from the failed memory device for future operations.

7. The method of claim 1, further comprising logging one or more of

an identity of a failed memory device;

and

memory location data specifying one or more storage regions of the failed memory device comprising one or more errors.

8. The method of claim 7, further comprising logging memory location data specifying one or more additional storage regions comprising one or more errors in the failed memory device in response to determining that the one or more additional storage regions comprise uncorrectable errors that are correctable using replacement data for the failed memory device.

9. The method of claim 1, further comprising replacing data in one or more storage regions in a failed memory device with replacement data from one or more extra memory devices for future operations such that data in the failed memory device outside the one or more storage regions is not replaced.

10. The method of claim 9, further comprising replacing data from one or more additional storage regions on one or more of a failed memory device and a different memory device with replacement data from the same one or more extra memory devices, each of the one or more additional storage regions comprising one or more errors, wherein the one or more storage regions and the one or more additional storage regions do not share a common logical page.

11. The method of claim 9, further comprising replacing one or more additional storage regions with replacement data from one or more different extra memory devices, each of the one or more additional storage regions comprising one or more errors, wherein the one or more storage regions and the one or more additional storage regions share a common logical page.

12. An apparatus for managing data storage, the apparatus comprising:

an ECC module configured to determine that an error correcting code (ECC) chunk comprises an uncorrectable number of errors, the ECC chunk stored across a plurality of storage regions;

an isolation module configured to test individual storage regions of the plurality of storage regions by iteratively substituting replacement data, within data of the ECC chunk, for individual storage regions of the plurality of storage regions to form substitute ECC chunks until one of the substitute ECC chunks is correctable using the error correcting code for the ECC chunk;

the isolation module configured to determine that a storage region that stored data for the ECC chunk is in error, wherein the isolation module replaces the stored data for the ECC chunk with the replacement data in the one of the substitute ECC chunks; and

wherein the ECC module and the isolation module comprise one or more of logic hardware and a non-transitory computer readable storage medium storing executable code.

13. The apparatus of claim 12, wherein the plurality of storage regions comprise at least a portion of an array of memory devices and wherein at least one of the memory devices of the array stores parity information derived from data stored on each memory device of the plurality of memory devices storing the ECC chunk, the replacement data derived from the parity information.

14. The apparatus of claim 12, further comprising a logging module configured to, in response to the isolation module determining that a storage region is in error, log memory location data specifying the storage region in error.

15. The apparatus of claim 14, wherein a storage region comprises one or more of one or more ECC chunks, one or more physical pages, one or more logical pages, one or more physical erase blocks, one or more logical erase blocks, a chip, a portion of a chip, a portion of one or more dies, and one or more dies.

16. The apparatus of claim 14, further comprising an analysis module configured to analyze a log of the logging module to determine one or more of an area in the storage region with data errors and an error type.

17. The apparatus of claim 12, further comprising an error storage recovery module configured to erase an erase block of the storage region in error and to prepare the erase block for future data storage in response to the ECC module determining that one of the substitute ECC chunks is correctable using the error correcting code for the ECC chunk, the erase block comprising a portion of the ECC chunk.

18. The apparatus of claim 12, further comprising a correct data module configured to return corrected data in response to the isolation module determining the storage region in error.

19. The apparatus of claim 12, further comprising a reporting module configured to report the storage region in error in response to the isolation module determining the storage region in error.

20. A system for managing data storage, the system comprising:

an array of non-volatile storage elements, the array comprising two or more non-volatile storage elements and one or more extra non-volatile storage elements, the extra non-volatile storage elements storing parity information from data stored on each non-volatile storage element of the two or more non-volatile storage elements; and

a storage controller configured to,

determine that an error correcting code (ECC) chunk has more errors than are correctable using an error correcting code for the ECC chunk, the ECC chunk stored across the two or more non-volatile storage elements;

iteratively substitute replacement data, within data of the ECC chunk, for individual non-volatile storage elements of the two or more non-volatile memory devices storage elements to form substitute ECC chunks until one of the substitute ECC chunks is correctable using the error correcting code for the ECC chunk, wherein the replacement data is derived from the parity information; and

provide corrected data from the correctable one of the substitute ECC chunks.

21. The system of claim 20, further comprising a computer and a solid-state storage device coupled to the computer over one or more communications buses, the solid-state storage device comprising the array of non-volatile storage elements.

22. A computer program product comprising a non-transitory computer readable storage medium storing computer usable program code executable to perform operations for managing data storage, the operations comprising:

determining that an error correcting code (ECC) block has more errors than are correctable using an error correcting code for the ECC block, the ECC block stored across a plurality of storage regions of an array of storage regions, wherein at least one of the storage regions of the array stores parity information derived from data stored on each storage region of the plurality of storage regions storing the ECC block;

iteratively substituting replacement data, within data of the ECC block, for individual storage regions of the plurality of storage regions to form substitute ECC blocks until one of the substitute ECC blocks is correctable using the error correcting code for the ECC block; and

providing corrected data from the correctable one of the substitute ECC blocks.

23. The computer program product of claim 22, the operations further comprising retiring one or more of the storage regions associated with the substitute ECC chunk, the one or more storage regions comprising one or more errors.

24. An apparatus for managing data storage, the apparatus comprising:

means for determining that an error correcting code (ECC) chunk has more errors than are correctable using an error correcting code for the ECC chunk, the ECC chunk stored across a plurality of storage elements, the error correcting code derived from data of the ECC chunk;

means for iteratively substituting replacement data, within data of the ECC chunk, for individual storage elements of the plurality of storage elements to form substitute ECC chunks until one of the substitute ECC chunks is correctable using the error correcting code for the ECC chunk; and

means for providing corrected data from the one of the substitute ECC chunks.

25. The apparatus of claim 24, further comprising means for replacing data in one or more storage regions in a failed storage element with replacement data from one or more extra storage elements for future operations such that data in the failed storage element outside the one or more storage regions is not replaced.