STORING CACHED DATA IN OVER-PROVISIONED MEMORY IN RESPONSE TO POWER LOSS

• US 20130191578A1
• Filed: 01/20/2012
• Published: 07/25/2013
• Est. Priority Date: 01/20/2012
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Inventors: Goss, Ryan J., Seekins, David S.
• Granted Patent: US 9,158,700 B2
• Time in Patent Office: Days
• US Class Current: 711/102
• CPC Class Codes: G06F 11/1441 Resetting or repowering
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 12/10 Address translation
• CPC Class Codes: G06F 2212/214 Solid state disk

Abstract

A power loss condition is detected that affects volatile data that is cached in preparation for storage in a non-volatile, solid-state memory device. The volatile cached data is stored in an over-provisioned portion of the non-volatile, solid-state memory device in response to the power loss condition.

Claims

1. A method comprising:

detecting a power loss condition affecting volatile data that is cached in preparation for storage in a non-volatile, solid-state memory device; and

storing the volatile cached data in an over-provisioned portion of the non-volatile, solid-state memory device in response to the power loss condition.

2. The method of claim 1, further comprising performing garbage collection on the over-provisioned portion of the non-volatile, solid-state memory device before storing the volatile cached data.

3. The method of claim 1, further comprising dynamically-allocating logical zones within the non-volatile, solid-state memory device, wherein the over-provisioned portion is selected from one of the dynamically-allocated, logical zones.

4. The method of claim 3, wherein the logical zones are differentiated based on any combination of retention, performance, and reliability characteristics of physical storage to which the logical zones are mapped.

5. The method of claim 1, wherein the non-volatile, solid-state memory device is acting as a cache for another non-volatile data storage device.

6. The method of claim 1, further comprising, upon resumption of power, moving the stored cached data from the over-provisioned portion of the non-volatile, solid-state memory device to thereafter free the over-provisioned portion for purposes of minimizing write-amplification.

7. The method of claim 6, wherein the cached data is moved to a volatile memory upon resumption of the power.

8. The method of claim 6, wherein the cached data is moved to a provisioned portion of the non-volatile, solid-state memory device upon resumption of the power.

9. An apparatus comprising:

a memory interface configured to access a non-volatile, solid-state memory having an over-provisioned portion; and

a controller configured to;

detect a power loss condition affecting the apparatus;

access cached data of a volatile memory that stores data in preparation for storage to the non-volatile, solid-state memory; and

store the cached data in the over-provisioned portion in response to the power loss condition.

10. The apparatus of claim 9, wherein the controller is further configured to perform garbage collection on the over-provisioned portion of the solid-state memory before storing the cached data.

11. The apparatus of claim 9, wherein the controller is further configured to dynamically-allocate logical zones within the non-volatile, solid-state memory, and wherein the over-provisioned portion is selected from one of the dynamically-allocated, logical zones.

12. The apparatus of claim 11, wherein the logical zones are differentiated based on any combination of retention, performance, and reliability characteristics of physical storage to which the logical zones are mapped.

13. The apparatus of claim 9, wherein the apparatus is acting as a cache for another non-volatile data storage device.

14. The apparatus of claim 9, wherein the controller is further configured to, upon resumption of power, move the stored cached data from the over-provisioned portion of the non-volatile, solid-state memory to thereafter free the over-provisioned portion for purposes of minimizing write-amplification.

15. The apparatus of claim 14, wherein the cached data is moved from the over-provisioned portion to the volatile memory upon resumption of the power.

16. The apparatus of claim 14, wherein the cached data is moved from the over-provisioned portion to the provisioned portion of the non-volatile, solid-state memory upon resumption of the power.

17-20. -20. (canceled)

21. A method comprising:

detecting an unexpected power loss condition affecting volatile data that is cached in preparation for storage in a non-volatile, solid-state memory device, wherein the non-volatile, solid-state memory device comprises advertised and over-provisioned portions, and wherein the non-volatile, solid-state memory device is acting as a cache for a lower tier, non-volatile data storage device;

determining that the advertised portion is unable to accept the volatile data in response to the power loss condition; and

in response thereto, storing the volatile cached data in an over-provisioned portion of the non-volatile, solid-state memory device.

22. The method of claim 21, further comprising performing garbage collection on the over-provisioned portion before storing the volatile cached data.

23. The method of claim 21, further comprising, upon resumption of power, moving the stored cached data from the over-provisioned portion of the non-volatile, solid-state memory to thereafter free the over-provisioned portion for purposes of minimizing write-amplification.

24. The method of claim 21, further comprising forgoing any movement of data from the non-volatile, solid-state memory device to the lower tier, non-volatile data storage device in response to the power loss condition.