Method and apparatus for using wear-out blocks in nonvolatile memory

• US 9,430,339 B1
• Filed: 12/24/2013
• Issued: 08/30/2016
• Est. Priority Date: 12/27/2012
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: Marvell International Limited (Marvell Technology Group Limited)
• Inventors: Song, Jong-uk
• Primary Examiner(s): BUTLER, SARAI E
• Application Number:
  US14/140,295
• Time in Patent Office: 980 Days
• Field of Search: 714/42
• US Class Current: 1/1
• CPC Class Codes: G11C 16/3418 Disturbance prevention or evaluation; Refreshing of disturbed memory data
• 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

Abstract

Some of the embodiments of the present disclosure provide a method comprising: storing volatile data in nonvolatile memory; over a time interval, periodically refreshing the volatile data by (i) reading the volatile data from the nonvolatile memory and (ii) rewriting the volatile data to the nonvolatile memory; determining a number of errors in the volatile data that is read during the periodic refresh of the volatile data; and based, at least in part, on the number of errors that is determined, modifying the time interval. The method may also comprise decreasing the time interval if the number of errors is determined to be greater than an error threshold value.

Claims

1. A method comprising:

designating a portion of a nonvolatile memory as a wear-out portion of the nonvolatile memory, in response to the portion of the nonvolatile memory being error-prone beyond a particular threshold;

storing volatile data in the wear-out portion of the nonvolatile memory;

over a time interval, periodically refreshing the volatile data in the wear-out portion of the nonvolatile memory by (i) reading the volatile data from the wear-out portion of the nonvolatile memory and (ii) rewriting the volatile data to the wear-out portion of the nonvolatile memory;

determining a number of errors in the volatile data that is read during the periodic refresh of the volatile data; and

based, at least in part, on the number of errors that is determined, modifying the time interval.

2. The method of claim 1, further comprising:

if the number of errors is determined to be greater than an error threshold value, decreasing the time interval.

3. The method of claim 1, the method further comprising:

if the time interval that is modified is less than a time threshold, discontinuing use of the wear-out portion of the nonvolatile memory for the volatile data.

4. The method of claim 3, further comprising:

writing the volatile data to another portion of the nonvolatile memory.

5. The method of claim 1, wherein:

the nonvolatile memory is managed by a memory controller that includes an error correction code; and

the nonvolatile memory includes memory cells that produce errors that are not correctable by the error correction code.

6. The method of claim 1, wherein determining the number of errors in the volatile data that is read during the periodic refresh of the volatile data is performed by an error correction code (i) after reading the volatile data from the wear-out portion of the nonvolatile memory and (ii) before rewriting the volatile data to the wear-out portion of the nonvolatile memory.

7. The method of claim 1, wherein the periodic refresh of the volatile data is performed without erasing any portion of the volatile data stored in the wear-out portion of the nonvolatile memory.

8. A system comprising:

a nonvolatile memory; and

a memory controller, wherein the memory controller is configured to

designate a portion of the nonvolatile memory as a wear-out portion of the nonvolatile memory, in response to the portion of the nonvolatile memory being error-prone beyond a particular threshold;

store volatile data in the wear-out portion of the nonvolatile memory,

over a time interval, periodically refresh the volatile data in the wear-out portion of the nonvolatile memory by (i) reading the volatile data from the wear-out portion of the nonvolatile memory and (ii) rewriting the volatile data to the wear-out portion of the nonvolatile memory,

determine if a number of errors in the volatile data that is read during the periodic refresh of the volatile data is greater than an error threshold value, and

modify the time interval based, at least in part, on the number of errors that is determined.

9. The system of claim 8, wherein the memory controller is further configured to:

discontinue use of the wear-out portion of the nonvolatile memory for the volatile data if the time interval that is modified is less than a time threshold.

10. The system of claim 9, wherein the memory controller is further configured to:

write the volatile data to another wear-out portion of the nonvolatile memory.

11. The system of claim 8, wherein

the memory controller includes an error correction code, and

the nonvolatile memory includes memory cells that produce errors that are not correctable by the error correction code.

12. The system of claim 8, wherein the memory controller is configured to determine the number of errors in the volatile data that is read during the periodic refresh of the volatile data by using an error correction code (i) after the volatile data is read from the wear-out portion of the nonvolatile memory and (ii) before the volatile data is rewritten to the wear-out portion of the nonvolatile memory.

13. The system of claim 8, wherein the periodic refresh of the volatile data is performed without erasing any portion of the volatile data stored in the wear-out portion of the nonvolatile memory.

14. A nonvolatile memory device comprising:

a memory cell array including a wear-out memory portion that is error-prone beyond a predetermined threshold; and

a controller configured to

store volatile data in the wear-out memory portion,

refresh the volatile data by (i) reading the volatile data from the wear-out memory portion and (ii) rewriting the volatile data to the wear-out memory portion, wherein the volatile data is refreshed periodically at a time interval,

determine if a number of errors in the volatile data that is read during the periodic refresh of the volatile data is greater than an error threshold value, and

modify the time interval based, at least in part, on the number of errors detected in the volatile data that is read during the periodic refresh.

15. The nonvolatile memory device of claim 14, wherein the controller is further configured to:

store nonvolatile data in the memory cell array in memory cells that are not included in the wear-out memory portion.

16. The nonvolatile memory device of claim 14, wherein the controller is further configured to:

discontinue use of the wear-out memory portion for the volatile data if the time interval that is modified is less than a time threshold.

17. The nonvolatile memory device of claim 14, wherein the controller is further configured to:

write the volatile data to another wear-out memory portion in the memory cell array.

18. The nonvolatile memory device of claim 14, wherein:

the controller includes an error correction code; and

the wear-out memory portion includes memory cells that produce errors that are not correctable by the error correction code.

19. The nonvolatile memory device of claim 14, wherein the controller is configured to determine the number of errors in the volatile data that is read during the periodic refresh of the volatile data by using an error correction code (i) after the volatile data is read from the wear-out memory portion and (ii) before the volatile data is rewritten to the wear-out memory portion.

20. The nonvolatile memory device of claim 14, wherein the periodic refresh of the volatile data is performed without erasing any portion of the wear-out memory portion.