Configurable accelerated post-write read to manage errors

• US 8,566,671 B1
• Filed: 06/29/2012
• Issued: 10/22/2013
• Est. Priority Date: 06/29/2012
• Status: Active Grant

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: Sandisk Technologies Incorporated (Western Digital Corporation)
• Inventors: Kochar, Mrinal, Ye, Jia, Huang, Jianmin
• Primary Examiner(s): Bhatia, Ajay
• Assistant Examiner(s): NGUYEN, THIEN DANG
• Application Number:
  US13/538,927
• Time in Patent Office: 480 Days
• Field of Search: 714/764
• US Class Current: 714/764
• CPC Class Codes: G06F 11/1048 using arrangements adapted for a specific error detection or correction feature
• CPC Class Codes: G11C 16/00 Erasable programmable read-only memories G11C14/00 takes precedence
• CPC Class Codes: G11C 2029/0409 Online test
• CPC Class Codes: G11C 2029/0411 Online error correction
• CPC Class Codes: G11C 29/42 using error correcting codes [ECC] or parity check

Abstract

Data errors in non-volatile memory inevitably increase with usage and with higher density of bits stored per cell. The memory is configured to have a first portion operating with less error but of lower density storage, and a second portion operating with a higher density but less robust storage. An error management provides reading and checking the copy after copying to the second portion. If the copy has excessive error bits, it is repeated in a different location either in the second or first portion. The reading and checking of the copy is accelerated by reading only a sample of it. The sample is selected from a subset of the copy having its own ECC, where the sample selected depends on the count of erase-program cycles that a block has experienced, where different count ranges can use different samples.

Claims

1. A method of operating a memory system including a non-volatile memory circuit formed of one or more blocks of non-volatile memory cells, the block being a minimum erase unit, where each of the blocks includes one or more groups of memory cells that can be operated in parallel, the method comprising:

for each of the blocks, maintaining by the memory system of an experience count indicative of the number of erase-program cycles that the corresponding block has experienced;

maintaining by the memory system of a file wherein, for at least first and second non-overlapping ranges of experience counts, a corresponding first and second sample of data in a group of previously programmed memory cells is to be selected for a post-write process, wherein the file is configurable so that the first and second samples can be configured independently;

programming multiple subsets of data into a first group of memory cells, each subset of data being provided with an ECC;

subsequent to said programming, selecting by the memory system of a sample of the data programmed in the first group of memory cells from said multiple subsets of data programmed into the first group, wherein the selecting of a sample includes;

determining whether the corresponding experience count for the first group of memory cells corresponds to one of the first or the second non-overlapping ranges of experience counts; and

selecting a sample based on the sample corresponding to the determined experience count in the file;

reading the selected sample;

checking the selected sample as read to determine an amount of error therein, and

reprogramming said multiple subsets of data into a second group of memory cells whenever the amount of error determined from the selected sample is more than a predetermined number of error bits.

2. The method as in claim 1, wherein:

each memory cell of the first group stores n bits of data;

said programming multiple subsets of data stores n data pages into the first group of memory cells, each data page constituted by one bit from each memory cell in the first group.

3. The method of claim 2, wherein the first and second samples share elements.

4. The method of claim 3, wherein the first sample is data corresponding to multiple pages from a first plurality of memory cells and the second sample is data corresponding to single page from the first plurality of memory cells.

5. The method of claim 4, wherein the second sample is data corresponding to the uppermost page of data from the first plurality of memory cells and the first sample is data corresponding the uppermost page of data and a middle page of data from the first plurality of memory cells.

6. The method of claim 2, wherein the first and second samples do not share any elements.

7. The method of claim 6, wherein the first sample is data corresponding to a first page from a first plurality of memory cells and the second sample is data corresponding to second page from the first plurality of memory cells, the first and second pages being different.

8. The method of claim 7, wherein the second sample is data corresponding to the uppermost page of data from the first plurality of memory cells.

9. The method as in claim 2, wherein n is equal to 3.

10. The method as in claim 1, wherein:

the first group of memory cells resides on a word line.

11. The method as in claim 1, wherein:

the first group forms a block.

12. The method as in claim 1, wherein:

said checking for errors include checking the ECC of the selected sample to determine the number of error bits.

13. The method as in claim 1, further comprising:

configuring said non-volatile memory into first and second portions, the first portion having memory cells operating with less error rate but lower density storage compared to the second portion; and

wherein;

said first and second groups of memory cells are in the first portion; and

said second group of memory cells is in the second portion.

14. The method as in claim 1, further comprising:

configuring said non-volatile memory into first and second portions, the first portion having memory cells operating with less error rate but lower density storage compared to the second portion; and

wherein;

said first group of memory cells is in the first portion; and

said second group of memory cells is in the second portion.

15. The method as in claim 1, wherein:

the memory is further organized into multiple rows of NAND chains;

each NAND chain being a column of a string of daisy-chained memory cells terminating at a source terminal and a drain terminal; and

the selected sample is a subset of data stored at memory cells closest to the source terminal of each NAND chain.

16. The method as in claim 1, wherein:

the memory is further organized into multiple rows of NAND chains;

each NAND chain being a column of a string of daisy-chained memory cells terminating at a source terminal and a drain terminal; and

the selected sample is a subset of data stored at memory cells closest to the drain terminal of each NAND chain.

17. A non-volatile memory system comprising:

a memory circuit formed of an array of one or more blocks of non-volatile memory cells, the block being a minimum erase unit, where each of the blocks includes one or more groups of memory cells;

programming circuitry connectable to the memory array to program multiple subsets of data into a first group of memory cells, each subset of data being provided with an ECC;

logic circuitry, where the logic circuitry is configured to perform processes including;

for each of the blocks, maintaining an experience count indicative of the number of erase-program cycles that the corresponding block has experienced;

maintaining of a file wherein, for at least first and second non-overlapping ranges of experience counts, a corresponding first and second sample of data in a group of previously programmed memory cells is to be selected for a post-write process, wherein the file is configurable so that the first and second samples can be configured independently; and

selecting a sample of the data programmed in the first group of memory cells, the sample being selected from a subset of data said multiple subsets of data programmed into the first group, where selecting a sample includes

determining whether the corresponding experience count for the first group of memory cells corresponds to one of the first or the second non-overlapping ranges of experience counts; and

selecting a sample by the logic circuitry based on the sample corresponding to the determined experience count in the file;

reading circuitry connectable to the memory array to read the selected sample; and

ECC decoding circuitry connectable to the read circuitry to check errors in the selected sample,

wherein the programming circuitry is further connectable to the memory to reprogram said multiple subsets of data into a second group of memory cells whenever the errors checked from the selected sample is more than a predetermined number of error bits.

18. The non-volatile memory system of claim 17, wherein each memory of the cells of the first group stores n bits of data and said programming circuitry programs multiple subsets of data stores n data pages into the first group of memory cells, each data page constituted by one bit from each memory cell in the first group.

19. The non-volatile memory system of claim 18, wherein the first and second samples share elements.

20. The non-volatile memory system of claim 19, wherein the first sample is data corresponding to multiple pages from a first plurality of memory cells and the second sample is data corresponding to single page from the first plurality of memory cells.

21. The non-volatile memory system of claim 20, wherein the second sample is data corresponding to the uppermost page of data from the first plurality of memory cells and the first sample is data corresponding to the uppermost page of data and a middle page of data from the first plurality of memory cells.

22. The non-volatile memory system of claim 18, wherein the first and second samples do not share any elements.

23. The non-volatile memory system of claim 22, wherein the first sample is data corresponding to a first page from a first plurality of memory cells and the second sample is data corresponding to second page from the first plurality of memory cells, the first and second pages being different.

24. The non-volatile memory system of claim 23, wherein the second sample is data corresponding to the uppermost page of data from the first plurality of memory cells.

25. The non-volatile memory system of claim 18, wherein n is equal to 3.

26. The non-volatile memory system of claim 17, wherein the first group of memory cells resides on a word line.

27. The non-volatile memory system of claim 17, wherein the first group forms a block.

28. The non-volatile memory system of claim 17, wherein said ECC decoder checks the ECC of the selected sample to determine the number of error bits.

29. The non-volatile memory system of claim 17, wherein the memory circuit is configured as first and second portions, the first portion having memory cells operating with less error rate but lower density storage compared to the second portion;

and wherein;

said first and second groups of memory cells are in the first portion; and

said second group of memory cells is in the second portion.

30. The non-volatile memory system of claim 17, wherein the memory circuit is configured as first and second portions, the first portion having memory cells operating with less error rate but lower density storage compared to the second portion;

and wherein;

said first group of memory cells is in the first portion; and

said second group of memory cells is in the second portion.

31. The non-volatile memory system of claim 17, wherein:

the memory is further organized into multiple rows of NAND chains;

each NAND chain being a column of a string of daisy-chained memory cells terminating at a source terminal and a drain terminal; and

the selected sample is a subset of data stored at memory cells closest to the source terminal of each NAND chain.

32. The non-volatile memory system of claim 17, wherein:

the memory is further organized into multiple rows of NAND chains;

each NAND chain being a column of a string of daisy-chained memory cells terminating at a source terminal and a drain terminal; and

the selected sample is a subset of data stored at memory cells closest to the drain terminal of each NAND chain.