Memory module and method for performing wear-leveling of memory module using remapping, link, and spare area tables

• US 8,275,928 B2
• Filed: 05/15/2008
• Issued: 09/25/2012
• Est. Priority Date: 05/15/2008
• Status: Active Grant

• Current Assignee: Silicon Motion Incorporated California (Silicon Motion)
• Original Assignee: Silicon Motion Incorporated California (Silicon Motion)
• Inventors: Lin, Chien-Cheng
• Primary Examiner(s): Savla, Arpan P.
• Application Number:
  US12/121,108
• Publication Number:
  US 20090287875A1
• Time in Patent Office: 1,594 Days
• Field of Search: 711/103, 711/154, 711/165, 711/170
• US Class Current: 711/103
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory

Abstract

The invention comprises a memory module capable of wear-leveling. In one embodiment, the memory module comprises a flash memory and a controller. The flash memory comprises a plurality of management units, wherein each of the management units comprises a plurality of blocks. The controller receives new data with a logical address managed by a first management unit selected from the management units, pops a first spare block from a spare area of the first management unit, determines whether an erase count of the first spare block is greater than a first threshold value, searches a second management unit selected from the management units for a replacing block with an erase count lower than a second threshold value when the erase count of the first spare block is greater than the first threshold value, and directs the first management unit and the second management unit to exchange the first spare block with the replacing block.

Claims

1. A memory module capable of wear-leveling, comprising:

a flash memory, comprising a plurality of management units, wherein each of the management units comprises a plurality of blocks; and

a controller, receiving new data with a logical address managed by a first management unit selected from the management units, popping a first spare block from a spare area of the first management unit, determining whether an erase count of the first spare block is greater than a first threshold value, searching a second management unit selected from the management units for a replacing block with an erase count lower than a second threshold value when the erase count of the first spare block is greater than the first threshold value, and directing the first management unit and the second management unit to exchange the first spare block and the replacing block;

wherein each of the management units has a remapping table storing a first corresponding relationship between a plurality of physical addresses and a plurality of indexes corresponding to all of the blocks of the management unit, and each of the management units further has a link table storing a second corresponding relationship between a plurality of logical addresses of stored data and a first portion of the indexes corresponding to the blocks holding the stored data, and a spare area table storing a second portion of the indexes corresponding to the blocks storing no data.

2. The memory module as claimed in claim 1, wherein when the erase count of the first spare block is greater than the first threshold value, the controller pops a second spare block from the spare area of the first management unit, writes the new data to the second spare block, and then modifies a link table of the first management unit with the second corresponding relationship between a logical address of the new data and an index corresponding to the second spare block.

3. The memory module as claimed in claim 1, wherein when the erase count of the first spare block is greater than the first threshold value, the controller determines whether the replacing block is a spare block, and when the replacing block is not a spare block, the controller writes data stored in the replacing block to the first spare block, and then erases the replacing block, thus directing the first management unit and the second management unit to exchange the first spare block and the replacing block.

4. The memory module as claimed in claim 1, wherein when the erase count of the first spare block is greater than the first threshold value, the controller replaces a physical address of the first spare block in the first corresponding relationship of a remapping table of the first management unit with a physical address of the replacing block, and then replaces a physical address of the replacing block in the first corresponding relationship of a remapping table of the second management unit with a physical address of the first spare block, thus directing the first management unit and the second management unit to exchange the first spare block and the replacing block.

5. The memory module as claimed in claim 1, wherein when the erase count of the first spare block is not greater than the first threshold value, the controller writes the new data to the first spare block, and modifies a link table of the first management unit with the second corresponding relationship between a logical address of the new data and an index corresponding to the first spare block.

6. The memory module as claimed in claim 1, wherein each of the management units also has an erase count table storing a third corresponding relationship between the plurality of indexes of the blocks and a plurality of erase counts of the blocks.

7. The memory module as claimed in claim 1, wherein the second threshold value is a half of the first threshold value.

8. The memory module as claimed in claim 1, wherein the flash memory is a NAND flash memory.

9. The memory module as claimed in claim 1, wherein when the erase counts of all blocks of the second management unit are greater than the second threshold value, the controller searches a plurality of blocks of a third management unit other than the second management unit for the replacing block.

10. A method for performing wear-leveling in a memory module, wherein the memory module comprises a flash memory comprising a plurality of management units, and each of the management units comprises a plurality of blocks, comprising:

receiving new data with a logical address managed by a first management unit selected from the management units;

popping a first spare block from a spare area of the first management unit;

determining whether an erase count of the first spare block is greater than a first threshold value;

searching a second management unit selected from the management units for a replacing block with an erase count lower than a second threshold value when the erase count of the first spare block is greater than the first threshold value; and

directing the first management unit and the second management unit to exchange the first spare block and the replacing block;

wherein each of the management units has a remapping table storing a first corresponding relationship between a plurality of physical addresses and a plurality of indexes corresponding to all of the blocks of the management unit, and each of the management units further has a link table storing a second corresponding relationship between a plurality of logical addresses of stored data and a first portion of the indexes corresponding to the blocks holding the stored data, and a spare area table storing a second portion of the indexes corresponding to the blocks storing no data.

11. The method as claimed in claim 10, wherein the method further comprises:

popping a second spare block from the spare area of the first management unit when the erase count of the first spare block is greater than the first threshold value;

writing the new data to the second spare block; and

modifying a link table of the first management unit with the second corresponding relationship between a logical address of the new data and an index corresponding to the second spare block.

12. The method as claimed in claim 10, wherein exchange of the first spare block and the replacing block comprises:

determining whether the replacing block is a spare block;

writing data stored in the replacing block to the first spare block when the replacing block is not a spare block; and

erasing the replacing block when the replacing block is not a spare block.

13. The method as claimed in claim 10, wherein exchange of the first spare block and the replacing block further comprises:

replacing a physical address of the first spare block in the first corresponding relationship of a remapping table of the first management unit with a physical address of the replacing block; and

replacing a physical address of the replacing block in the first corresponding relationship of a remapping table of the second management unit with a physical address of the first spare block.

14. The method as claimed in claim 10, wherein the method further comprises:

when the erase count of the first spare block is not greater than the first threshold value, writing the new data to the first spare block; and

modifying a link table of the first management unit with the second corresponding relationship between a logical address of the new data and an index corresponding to the first spare block.

15. The method as claimed in claim 10, wherein each of the management units also has an erase count table storing a third corresponding relationship between the plurality of indexes of the blocks and a plurality of erase counts of the blocks.

16. The method as claimed in claim 10, wherein the second threshold value is a half of the first threshold value.

17. The method as claimed in claim 10, wherein the flash memory is a NAND flash memory.

18. The method as claimed in claim 10, wherein the method further comprises searching a plurality of blocks of a third management unit other than the second management unit for the replacing block when the erase counts of all blocks of the second management unit are greater than the second threshold value.