APPARATUS, SYSTEM, AND METHOD FOR PREDICTING FAILURES IN SOLID-STATE STORAGE

• US 20100122148A1
• Filed: 11/10/2009
• Published: 05/13/2010
• Est. Priority Date: 11/10/2008
• Status: Active Grant

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: Fusion-IO Incorporated (Western Digital Corporation)
• Inventors: Shober, Edward, Flynn, David, Thatcher, Jonathan
• Granted Patent: US 8,516,343 B2
• Time in Patent Office: Days
• US Class Current: 714/773
• 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 11/141 for bus or memory accesses
• 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 16/04 using variable threshold transistors, e.g. FAMOS
• CPC Class Codes: G11C 16/3418 Disturbance prevention or evaluation; Refreshing of disturbed memory data
• CPC Class Codes: G11C 16/3422 Circuits or methods to evaluate read or write disturbance in nonvolatile memory, without steps to mitigate the problem
• CPC Class Codes: G11C 16/349 Arrangements for evaluating degradation, retention or wearout, e.g. by counting erase cycles
• CPC Class Codes: G11C 2029/0411 Online error correction
• CPC Class Codes: G11C 2029/1208 Error catch memory
• CPC Class Codes: G11C 29/42 using error correcting codes [ECC] or parity check
• CPC Class Codes: G11C 29/82 for EEPROMs
• CPC Class Codes: G11C 29/832 with disconnection of faulty elements
• CPC Class Codes: G11C 29/88 with partially good memories

Abstract

An apparatus, system, and method are disclosed for predicting failures in solid-state storage and include a determination module a threshold module, a storage region error module, and a retirement module. The determination module determines that data stored in an ECC chunk contains Error Correcting Code (“ECC”) correctable errors and further determines a bit error count for the ECC chunk. The ECC chunk originates from non-volatile solid-state storage media. The threshold module determines that the bit error count satisfies an ECC chunk error threshold. The storage region error module determines that a storage region that contained contains at least a portion of the ECC chunk satisfies a region retirement criteria. The retirement module retires the storage region that contains at least a portion of the ECC chunk where the storage region satisfies the region retirement criteria.

Claims

1. An apparatus to determine non-volatile solid-state storage media status, the apparatus comprising:

a determination module that determines that data stored in an ECC chunk contains Error Correcting Code (“

ECC”

) correctable errors and further determines a bit error count for the ECC chunk, the ECC chunk originating from non-volatile solid-state storage media;

a threshold module that determines that the bit error count satisfies an ECC chunk error threshold;

a storage region error module that determines that a storage region that contains at least a portion of the ECC chunk satisfies a region retirement criteria; and

a retirement module that retires the storage region that contains at least a portion of the ECC chunk, the storage region satisfying the region retirement criteria.

2. The apparatus of claim 1, wherein the ECC chunk error threshold comprises a bit error count that satisfies an acceptable error threshold and is below a maximum number of ECC correctable errors for the ECC chunk.

3. The apparatus of claim 1, further comprising an error logging module that records an identifier associated with ECC chunk having the bit error count that satisfies the ECC chunk error threshold and a time indicator associated with determining the bit error count.

4. The apparatus of claim 1, further comprising

a media error prediction module that monitors a storage region retirement rate;

and

a retirement limit module that determines that the storage region retirement rate satisfies a storage region retirement threshold and sends a notification when the storage region retirement rate satisfies the storage region retirement threshold.

5. The apparatus of claim 1, wherein the storage region retired by the retirement module spans media elements of an array of solid-state storage elements such that each element includes a portion of the storage region and the portions of the storage region are accessed in parallel.

6. The apparatus of claim 1, wherein

the solid-state storage media comprises a plurality of solid-state memory elements arranged in two-dimensional array, each element comprising a plurality of physical erase blocks (“

PEBs”

), a group of PEBs comprising a logical erase block (“

LEB”

) wherein each LEB comprises at least one PEB from each solid-state memory element;

data is stored in a logical page, the logical page comprising a physical page in each PEB within an LEB; and

an ECC chunk comprises at least a portion of data from each physical page in a logical page.

7. The apparatus of claim 1, wherein the storage region comprises one or more of an ECC chunk, a plurality of ECC chunks, a page, a logical page spanning a plurality of solid-state storage chips, a PEB, a LEB, a chip, a row of chips, a column of chips, and a die on a chip.

8. The apparatus of claim 1, wherein the region retirement criteria further comprises determining that a storage region satisfies a region retirement criteria based on one or more data errors associated with the storage region wherein the one or more data errors are detected during one or more of a read operation, an erase operation, and a program operation.

9. The apparatus of claim 1, further comprising a verification module that verifies that the storage region retired by the retirement module is unreliable for storing data.

10. The apparatus of claim 9, wherein the verification module allows data storage in the storage region subsequent to the storage region error module determining that the storage region satisfies the region retirement criteria.

11. The apparatus of claim 1, further comprising a disturb counter that determines one or more of a read count and a program count for the storage region, the storage region error module determining that the storage region fails to satisfy the region retirement criteria in response to one or more of a read count and a write count satisfying a disturb threshold.

12. The apparatus of claim 11, wherein the storage region error module marks the storage region having one of a read count and a program count that satisfies the disturb threshold.

13. The apparatus of claim 1 further comprising a solid-state storage device comprising the non-volatile solid-state media.

14. The apparatus of claim 13, further comprising a computer in communication with the solid-state storage device over one of a bus and a computer network.

15. A method for determining solid-state storage device status, the method comprising:

determining that data stored in an ECC chunk contains Error Correcting Code (“

ECC”

) correctable errors;

determining a bit error count for the ECC chunk, the ECC chunk originating from non-volatile solid-state storage media;

determining that the bit error count satisfies an ECC chunk error threshold;

determining that a storage region that contains at least a portion of the ECC chunk satisfies a region retirement criteria; and

retiring the storage region that contains at least a portion of the ECC chunk, the storage region satisfying the region retirement criteria.

16. The method of claim 15, wherein the ECC chunk error threshold comprises a bit error count that satisfies an acceptable error threshold and is below a maximum number of ECC correctable errors for the ECC chunk.

17. The method of claim 15, further comprising recording a time indicator associated with determining the bit error count and recording an identifier for the ECC chunk having the bit error count that satisfied the ECC chunk error threshold.

18. The method of claim 15, further comprising determining that a storage region retirement rate satisfies a storage region retirement threshold, the storage region retirement rate determined from recorded time indicators associated with each retirement of a storage region.

19. The method of claim 15, incrementing a disturb counter for a storage region for one or more of each read operation and each program operation of the storage region, wherein a retirement region satisfying a region retirement criteria is not retired in response to the disturb counter for the storage region satisfying a disturb threshold.

20. An apparatus for retiring a storage region, the apparatus comprising:

an error module that determines that a storage region has one or more data errors, the storage region comprising non-volatile solid-state storage, the error module comprising

an Error Correcting Code (“

ECC”

) error module that determines that data stored in an ECC chunk contains errors that are correctable with ECC an associated ECC algorithm and further determines a bit error count for the ECC chunk, the ECC chunk originating from a storage region in non-volatile solid-state storage media;

a threshold module that determines that the ECC chunk has a data error by determining that the bit error count satisfies an ECC chunk error threshold;

an erase error module that detects a data error in the storage region during an erase operation; and

a program error module that detects a data error in the storage region during a program operation;

a storage region error module that determines that a storage region determined to have one or more data errors by the error module satisfies a region retirement criteria; and

a retirement module that retires the storage region, the storage region satisfying the region retirement criteria.

21. The apparatus of claim 20, further comprising a retry module that verifies that a retired storage region has one or more data errors, the storage region originally retired by retirement module based on a reported hardware error for the storage region.

22. The apparatus of claim 21, wherein the retry module determines if a storage region has one or more data errors by using a garbage collection operation, the garbage collection operation comprising at least performing an erase operation on the storage region and detecting one or more data errors during the erase operation.