Systems and methods for multi-zone data tiering for endurance extension in solid state drives

• US 10,691,531 B2
• Filed: 12/04/2014
• Issued: 06/23/2020
• Est. Priority Date: 12/04/2014
• Status: Active Grant

• Current Assignee: JP Morgan Chase Bank, N.A. (JP Morgan Chase & Co)
• Original Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Inventors: Singhai, Ashish, Karamcheti, Vijay, Narasimha, Ashwin
• Primary Examiner(s): Knapp, Justin R
• Application Number:
  US14/560,802
• Publication Number:
  US 20160162354A1
• Time in Patent Office: 2,028 Days
• Field of Search: None
• CPC Class Codes: G06F 11/1048 using arrangements adapted for a specific error detection or correction feature
• CPC Class Codes: G11C 2029/0411 Online error correction
• CPC Class Codes: G11C 29/028 with adaption or trimming of parameters
• CPC Class Codes: G11C 29/52 Protection of memory contents; Detection of errors in memory contents
• CPC Class Codes: H03M 13/1102 Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes
• CPC Class Codes: H03M 13/152 Bose-Chaudhuri-Hocquenghem [BCH] codes
• CPC Class Codes: H03M 13/353 Adaptation to the channel
• CPC Class Codes: H03M 13/356 Unequal error protection [UEP]

Abstract

Systems and methods for increasing the endurance of a solid state drive are disclosed. The disclosed systems and methods can assign different levels of error protection to a plurality of blocks of the solid state drive. The disclosed methods can provide a plurality of error correction mechanisms, each having a plurality of corresponding error correction levels and associate a first plurality of blocks of the solid state drive with a first zone and a second plurality of blocks of the solid state drive with a second zone. The disclosed methods can assign a first error correction mechanism and a first corresponding error correction level to the first zone and can assign a second error correction mechanism and a second corresponding error correction level to the second zone.

Claims

1. A method, comprising:

providing a plurality of error correction mechanisms, each having a plurality of corresponding error correction levels;

associating a first plurality of blocks of a solid state drive with a first zone, wherein the first zone comprises a first logical accumulation of blocks;

associating a second plurality of blocks of the solid state drive with a second zone, wherein the second zone comprises a second logical accumulation of blocks, and wherein the first zone is configured to support a larger number of program/erase (PE) cycles compared to the second zone by limiting programming to lower pages in the first zone;

assigning a first error correction mechanism and a first corresponding error correction level to the first zone;

assigning a second error correction mechanism and a second corresponding error correction level to the second zone;

directing a first plurality of write requests to the solid state drive into the first zone and a second plurality of write requests into the second zone, wherein the first plurality of write requests is for data that is overwritten more frequently than data for the second plurality of write requests; and

re-directing at least one write request from the first plurality of write requests into the second zone.

2. The method of claim 1, further comprising re-associating a first block from the first plurality of blocks with the second zone, if the first block has an error count number higher than a threshold error count number.

3. The method of claim 1, further comprising re-associating a second block from the second plurality of blocks with the first zone, if the second block has an error count number lower than a threshold error count number.

4. The method of claim 1, further comprising re-assigning to the first zone the first error correction mechanism and a third corresponding error correction level, wherein the first error correction mechanism and the third corresponding error correction level provide higher error correction from the first error correction mechanism and the first corresponding error correction level.

5. The method of claim 1, further comprising re-assigning to the first zone a third error correction mechanism and the first corresponding error correction level, wherein the third error correction mechanism and the first corresponding error correction level provide higher error correction from the first error correction mechanism and the first corresponding error correction level.

6. The method of claim 1,

wherein write requests of the first plurality of write requests generate higher write amplification than write requests of the second plurality of write requests.

7. The method of claim 1, wherein the first zone covers at least 10% of a capacity of the solid state drive.

8. The method of claim 1, further comprising filtering out write requests with random traffic patterns from the second zone.

9. The method of claim 1, wherein the solid state drive is a multi-level cell flash device.

10. The method of claim 9, further comprising dividing the multi-level cell flash device into at least one region, wherein the at least one region is in a single-level cell mode.

11. The method of claim 10, wherein the first zone is assigned to the at least one region.

12. A memory controller, comprising:

a controller module configured to;

communicate with a solid state drive having a plurality of blocks;

provide a plurality of error correction mechanisms, each having a plurality of corresponding error correction levels;

associate a first plurality of blocks of the solid state drive with a first zone, wherein the first zone comprises a first logical accumulation of blocks;

associate a second plurality of blocks of the solid state drive with a second zone, wherein the second zone comprises a second logical accumulation of blocks, and wherein the first zone is configured to support a larger number of program/erase (PE) cycles compared to the second zone by limiting programming to lower pages in the first zone;

assign a first error correction mechanism and a first corresponding error correction level to the first zone;

assign a second error correction mechanism and a second corresponding error correction level to the second zone;

direct a first plurality of write requests to the solid state drive into the first zone and a second plurality of write requests into the second zone, wherein the first plurality of write requests is for data that is overwritten more frequently than data for the second plurality of write requests; and

filter out write requests with random traffic patterns from the second zone.

13. The memory controller of claim 12, wherein the controller module is further configured to re-associate a first block from the first plurality of blocks with the second zone, if the first block has an error count number higher than a threshold error count number.

14. The memory controller of claim 12, wherein the controller module is further configured to re-associate a second block from the second plurality of blocks with the first zone, if the second block has an error count number lower than a threshold error count number.

15. The memory controller of claim 12, wherein the controller module is further configured to re-assign to the first zone the first error correction mechanism and a third corresponding error correction level, wherein the first error correction mechanism and the third corresponding error correction level provide higher error correction from the first error correction mechanism and the first corresponding error correction level.

16. The memory controller of claim 12, wherein the controller module is further configured to re-assign to the first zone a third error correction mechanism and the first corresponding error correction level, wherein the third error correction mechanism and the first corresponding error correction level provide higher error correction from the first error correction mechanism and the first corresponding error correction level.

17. The memory controller of claim 12,

wherein write requests of the first plurality of write requests are to generate higher write amplification than write requests of the second plurality of write requests.

18. The memory controller of claim 17, wherein the controller module is further configured to re-direct at least one write request from the first plurality of write requests into the second zone.

19. The memory controller of claim 12, wherein the first zone covers at least 10% of a capacity of the solid state drive.

20. The memory controller of claim 12, wherein the solid state drive is a multi-level cell flash device.

21. The memory controller of claim 20, wherein the controller module is further configured to divide the multi-level cell flash device into at least one region, wherein the at least one region is in a single-level cell mode.

22. The memory controller of claim 21, wherein the first zone is assigned to the at least one region.

23. An apparatus, comprising:

means for providing a plurality of error correction mechanisms, each having a plurality of corresponding error correction levels;

means for associating a first plurality of blocks of a solid state drive with a first zone, wherein the first zone comprises a first logical accumulation of blocks;

means for associating a second plurality of blocks of the solid state drive with a second zone, wherein the second zone comprises a second logical accumulation of blocks, and wherein the first zone is configured to support a larger number of program/erase (PE) cycles compared to the second zone by limiting programming to lower pages in the first zone;

means for assigning a first error correction mechanism and a first corresponding error correction level to the first zone;

means for assigning a second error correction mechanism and a second corresponding error correction level to the second zone;

means for directing a first plurality of write requests to the solid state drive into the first zone and a second plurality of write requests into the second zone, wherein the first plurality of write requests is for data that is overwritten more frequently than data for the second plurality of write requests;

means for associating a third plurality of blocks of the solid state drive with a third zone, wherein the third zone comprises a third logical accumulation of blocks;

means for assigning the second error correction mechanism and the second corresponding error correction level to the third zone;

means for receiving a third plurality of write requests, wherein the third plurality of write requests is for data that is overwritten more frequently than data for the second plurality of write requests;

in response to receiving the third plurality of write requests, means for changing an error correction mechanism and an error correction level of the third zone by assigning the first error correction mechanism and the first corresponding error correction level to the third zone; and

means for directing the third plurality of write requests into the third zone.