Methods, systems, and computer readable media for early detection of potential flash failures using an adaptive system level algorithm based on flash program verify

• US 9,349,476 B2
• Filed: 02/21/2013
• Issued: 05/24/2016
• Est. Priority Date: 02/21/2013
• Status: Active Grant

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: Sandisk Technologies Incorporated (Western Digital Corporation)
• Inventors: Pe'er, Assaf
• Primary Examiner(s): SCHELL, JOSEPH O
• Application Number:
  US13/773,570
• Publication Number:
  US 20140237298A1
• Time in Patent Office: 1,188 Days
• US Class Current: 1/1
• CPC Class Codes: G11C 11/5628 Programming or writing circuits; Data input circuits
• CPC Class Codes: G11C 16/00 Erasable programmable read-only memories G11C14/00 takes precedence
• CPC Class Codes: G11C 16/3418 Disturbance prevention or evaluation; Refreshing of disturbed memory data
• CPC Class Codes: G11C 16/349 Arrangements for evaluating degradation, retention or wearout, e.g. by counting erase cycles
• CPC Class Codes: G11C 16/3495 Circuits or methods to detect or delay wearout of nonvolatile EPROM or EEPROM memory devices, e.g. by counting numbers of erase or reprogram cycles, by using multiple memory areas serially or cyclically
• CPC Class Codes: G11C 2029/0409 Online test
• CPC Class Codes: G11C 2211/5621 Multilevel programming verification
• CPC Class Codes: G11C 29/021 in voltage or current generators
• CPC Class Codes: G11C 29/028 with adaption or trimming of parameters

Abstract

Methods, systems, and computer readable media for early detection of potential flash failures using an adaptive system level algorithm based on NAND program verify are disclosed. According to one aspect, a method for early detection of potential flash failures using an adaptive system level algorithm based on NAND program verify includes performing a program verify operation after a write to a non-volatile memory, where the program verify mechanism reports a pass or fail based on an existing measurement threshold value, and dynamically adjusting the measurement threshold value used by subsequent program verify operations based on the results of previous program verify operations.

Claims

1. A method for early detection of potential flash failures using an adaptive system level algorithm based on NAND program verify, the method comprising:

performing a program verify operation after a write to a page within a block of a non-volatile memory, wherein the program verify operation reports a pass or failure based on whether a number of bits failing to be written correctly from a buffer to the page in the non-volatile memory exceeds an existing measurement threshold value; and

dynamically adjusting the existing measurement threshold value used by subsequent program verify operations based on results of previous program verify operations, wherein dynamically adjusting the existing measurement threshold value includes determining that the program verify operation has reported the failure, and, in response to determining that the program verify operation has reported the failure;

determining whether the failure is recoverable;

in response to determining that the failure is recoverable, determining whether a number of recoverable failures for the block has exceeded a first program verify limit; and

in response to determining that the number of recoverable failures for the block has exceeded a first program verify limit, adjusting the existing measurement threshold value.

2. The method of claim 1 wherein the non-volatile memory comprises memory cells organized into at least one page, at least one page organized into at least one block, and at least one block occupying at least one die.

3. The method of claim 2 wherein dynamically adjusting the measurement threshold value comprises at least one of:

dynamically adjusting a threshold per page;

dynamically adjusting a threshold per block; and

dynamically adjusting a threshold per die.

4. The method of claim 1 wherein determining whether the failure is recoverable comprises reading data associated with the write and attempting to perform an error correction operation, wherein a successful error correction operation indicates that the failure is recoverable and an unsuccessful error correction operation indicates that the failure is not recoverable.

5. The method of claim 1 comprising, in response to determining that the number of recoverable failures has exceeded a first program verify limit, performing a data scrub to refresh the data stored within a portion of memory that contains the detected failure.

6. The method of claim 1 wherein determining whether the number of recoverable failures has exceeded a first program verify limit comprises determining whether the number of recoverable failures during a specified measurement period has exceeded a threshold failure rate.

7. The method of claim 1 comprising, in response to adjusting the measurement threshold, determining whether the adjusted measurement threshold exceeds a first measurement threshold limit, and, in response to determining that the adjusted measurement threshold exceeds a first measurement threshold limit, starting to count the number of program verify failures reported for a portion of the memory, wherein, upon a detection that the number of program verify failures reported for the portion of the memory exceeds a second program verify limit, the portion of the memory is marked as unusable.

8. The method of claim 7 comprising, in response to adjusting the measurement threshold, determining whether the adjusted measurement threshold exceeds a second measurement threshold limit, and, in response to determining that the adjusted measurement threshold exceeds a second measurement threshold limit, performing at least one of:

disabling the program verify operation for at least a portion of the non-volatile memory;

disabling write access to at least a portion of the non-volatile memory; and

disabling read access to at least a portion of the non-volatile memory.

9. The method of claim 1 wherein adjusting the measurement threshold comprises adjusting the measurement threshold to adjust sensitivity of the program verify method.

10. The method of claim 1 comprising, in response to determining that the failure is not recoverable, marking a portion of the memory that contains failing cells as unusable.

11. The method of claim 1 comprising saving the data associated with the write to a different portion of memory not marked as unusable.

12. The method of claim 1 further comprising:

identifying a portion of the memory for which the program verify operation has reported a fail;

copying data from the identified portion to another portion of the memory; and

storing the data associated with the write in the other portion of the memory.

13. The method of claim 1 wherein the program verify operation reports a pass or fail based on a number of memory cells that were not successfully programmed during the write.

14. The method of claim 13 wherein the program verify operation reports a fail when the number of memory cells that were not successfully programmed during the write exceeds the measurement threshold value.

15. The method of claim 14 wherein performing a program verify operation comprises performing a NAND erratic programming detection (EPD) and wherein the measurement threshold value comprises a bit scan program failure (BSPF) value.

16. The method of claim 1 wherein the program verify operation reports a pass or fail based on a measured programmed voltage value of a cell that was a target of the write.

17. The method of claim 16 wherein dynamically adjusting the measurement threshold value used by subsequent program verify operations comprises dynamically adjusting a voltage threshold used to determine whether the cell was successfully programmed or not.

18. The method of claim 1 wherein the program verify operation reports a pass or fail based on a distribution of measured programmed voltage values of cells that were targets of the write.

19. The method of claim 18 wherein dynamically adjusting the measurement threshold value used by subsequent program verify operations comprises dynamically adjusting measurement threshold values used to characterize the distribution of measured programmed voltage values of cells that were targets of the write.

20. The method of claim 13 wherein the program verify circuit reports a fail when the number of memory cells that were not successfully programmed during the write exceeds a measurement threshold value.

21. The method of claim 14 wherein the program verify circuit performs a NAND erratic programming detection (EPD) and wherein the measurement threshold value comprises a bit scan program failure (BSPF) value.

22. A system for early detection of potential flash failures using an adaptive system level algorithm based on NAND program verify, the system comprising:

a non-volatile memory, comprising a plurality of memory cells;

a program verify circuit for performing a program verify operation after a write to a page within a block of the non-volatile memory, wherein the program verify operation reports a pass or failure based on whether a number of bits failing to be written correctly from a buffer to the page in the non-volatile memory exceeds an existing measurement threshold value; and

a control circuit for dynamically adjusting the existing measurement threshold value used by subsequent program verify operations based on results of previous program verify operations by determining that the program verify operation has reported the failure, and, in response to determining that the program verify operation has reported the failure;

determining whether the failure is recoverable;

in response to determining that the failure is recoverable, determining whether a number of recoverable failures for the block has exceeded a first program verify limit; and

in response to determining that the number of recoverable failures for the block has exceeded a first program verify limit, adjusting the existing measurement threshold value.

23. The system of claim 22 wherein the non-volatile memory comprises memory cells organized into at least one page, at least one page organized into at least one block, and at least one block occupying at least one die.

24. The system of claim 23 wherein the control circuit dynamically adjusts at least one of:

a measurement threshold value per page;

a measurement threshold value per block; and

a measurement threshold value per die.

25. The system of claim 22 wherein determining whether the failure is recoverable comprises reading data associated with the write and attempting to perform an error correction operation, wherein a successful error correction operation indicates that the failure is recoverable and an unsuccessful error correction operation indicates that the failure is not recoverable.

26. The system of claim 22 wherein, in response to determining that the number of recoverable failures has exceeded a first program verify limit, a data scrub to refresh the data stored within a portion of memory that contains the detected failure is performed.

27. The system of claim 22 wherein determining whether the number of recoverable failures has exceeded a first program verify limit comprises determining whether the number of recoverable failures during a specified measurement period has exceeded a threshold failure rate.

28. The system of claim 22 wherein the control circuit, in response to adjusting the measurement threshold, determines whether the adjusted measurement threshold exceeds a first measurement threshold limit, and, in response to determining that the adjusted measurement threshold exceeds a first measurement threshold limit, starts to count the number of program verify failures reported for a portion of the memory, wherein, upon a detection that the number of program verify failures reported for the portion of the memory exceeds a second program verify limit, the portion of the memory is marked as unusable.

29. The system of claim 28 comprising, in response to adjusting the measurement threshold, determining whether the adjusted measurement threshold exceeds a second measurement threshold limit, and, in response to determining that the adjusted measurement threshold exceeds a second measurement threshold limit, performing at least one of:

disabling the program verify operation for at least a portion of the non-volatile memory;

disabling write access to at least a portion of the non-volatile memory; and

disabling read access to at least a portion of the non-volatile memory.

30. The system of claim 22 wherein adjusting the measurement threshold comprises adjusting the measurement threshold to adjust sensitivity of the program verify method.

31. The system of claim 22 wherein the control circuit, in response to determining that the failure is not recoverable, marks a portion of the memory that contains failing cells as unusable.

32. The system of claim 31 wherein the control circuit saves the data associated with the write to a different portion of memory not marked as unusable.

33. The system of claim 22 wherein the control circuit is configured to:

identify a portion of the memory for which the program verify operation has reported a fail;

copy data from the identified portion to another portion of the memory; and

store the data associated with the write in the other portion of the memory.

34. The system of claim 22 wherein the program verify circuit reports a pass or fail based on a number of memory cells that were not successfully programmed during the write.

35. The system of claim 22 wherein the program verify circuit reports a pass or fail based on a measured programmed voltage value of a cell that was a target of the write.

36. The system of claim 35 wherein dynamically adjusting the measurement threshold value used by subsequent program verify operations comprises dynamically adjusting a voltage threshold used to determine whether the cell was successfully programmed or not.

37. The system of claim 22 wherein the program verify circuit reports a pass or fail based on a distribution of measured programmed voltage values of cells that were targets of the write.

38. The system of claim 37 wherein dynamically adjusting the measurement threshold value used by subsequent program verify operations comprises dynamically adjusting measurement threshold values used to characterize the distribution of measured programmed voltage values of cells that were targets of the write.

39. A non-transitory computer readable medium having stored thereon executable instructions that when executed by the processor of a computer control the computer to perform steps comprising:

performing a program verify operation after a write to a page within a block of a non-volatile memory, wherein the program verify operation reports a pass or failure based on whether a number of bits failing to be written correctly from a buffer to the page in the non-volatile memory exceeds an existing measurement threshold value; and

dynamically adjusting the existing measurement threshold value used by subsequent program verify operations based on results of previous program verify operations, wherein dynamically adjusting the existing measurement threshold value includes determining that the program verify operation has reported the failure, and, in response to determining that the program verify operation has reported the failure;

determining whether the failure is recoverable;

in response to determining that the failure is recoverable, determining whether a number of recoverable failures for the block has exceeded a first program verify limit; and

in response to determining that the number of recoverable failures for the block has exceeded a first program verify limit, adjusting the existing measurement threshold value.