Wear-leveling of memory devices

• US 10,061,524 B2
• Filed: 09/01/2016
• Issued: 08/28/2018
• Est. Priority Date: 09/01/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Baptist, Andrew D., Resch, Jason K., Volvovski, Ilya
• Primary Examiner(s): Farrokh, Hashem
• Application Number:
  US15/254,294
• Publication Number:
  US 20180059962A1
• Time in Patent Office: 726 Days
• CPC Class Codes: G06F 3/0616 in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A method for execution by one or more processing modules of one or more computing devices of a dispersed storage network (DSN) begins by tracking a number of reads and writes to a memory device within DSN memory. The method continues by mapping the tracked number of reads and writes to a counter that tracks a physical region of the memory device that is impacted by the reads and writes. The method continues by determining a highest-to-lowest ranking of memory device wear based on the mapping. The method continues by moving data located at a highly ranked data location to a new location with a lower ranking.

Claims

1. A method for execution by one or more processing modules of one or more computing devices of a dispersed storage network (DSN), the method comprises:

tracking a number of reads and writes to a memory device within DSN memory;

mapping the tracked number of reads and writes to a counter that tracks a physical region of the memory device that is impacted by the reads and writes;

determining a highest-to-lowest ranking of memory device wear based on the mapping; and

moving data located at a highly ranked data location to a new location with a lower ranking; and

wherein the moving data includes at least a memory device log moved to the new location with a lower ranking and further includes any of;

re-writing the memory device log or writing the memory device log to an explicit position in the memory device.

2. The method of claim 1, wherein the lower ranking is a lowest ranking.

3. The method of claim 1, wherein the memory device log is included in any of:

a file, an offset within the memory device, a zone in a SMR (Shingled Magnetic Recording) drive, a chip within an SSD (Solid State Drive), or a position in a tape reel.

4. The method of claim 1, wherein the tracked number of reads and writes is based on any of:

bytes, slices, blocks, or time spent reading and/or writing.

5. The method of claim 1, wherein the memory device internally monitors the tracked number of reads and writes.

6. The method of claim 1, wherein the tracked number of reads and writes includes any of:

SMR (Shingled Magnetic Recording) drive tracking IO (Input/Output) operations or bytes written or read to each zone.

7. The method of claim 1, wherein the memory device includes any of:

encoded data slices storage, Packed Slice Storage (PSS), and or Zone Slice Storage (ZSS).

8. The method of claim 1, wherein the data is encoded data slices.

9. A processing unit within a dispersed storage network (DSN) comprises:

an interface;

a local memory; and

a processing module operably coupled to the interface and the local memory, wherein the processing module functions to;

track a number of reads and writes to a memory device within DSN memory;

map the tracked number of reads and writes to a counter which tracks a physical region of the memory device that is impacted by the reads and writes;

determine a highest-to-lowest ranking of memory device wear based on the mapping; and

when a memory device log is re-written, move the memory device log to a location with a lower ranking; and

for encoded data slices stored in the memory device, rewrite highly ranked encoded data slices to a new location of the memory device based on the highest-to-lowest ranking of memory device wear.

10. The processing unit of claim 9, wherein the lower ranking is a lowest ranking.

11. The processing unit of claim 9, wherein the tracked number of reads and writes is based on any of:

bytes, slices, blocks, or time spent reading and/or writing.

12. The processing unit of claim 9, wherein the memory device internally monitors the tracked number of reads and writes.

13. The processing unit of claim 9, wherein the tracked number of reads and writes includes any of:

SMR (Shingled Magnetic Recording) drive tracking IO (Input/Output) operations or bytes written or read to each zone.

14. The processing unit of claim 9, wherein the memory device includes any of:

encoded data slice storage, Packed Slice Storage (PSS), and or Zone Slice Storage (ZSS).

15. A method for execution by one or more processing modules of one or more computing devices of a dispersed storage network (DSN), the method comprises:

tracking a number of reads and writes to a memory device within DSN memory, wherein the tracked number of reads and writes includes any of;

SMR (Shingled Magnetic Recording) drive tracking IO (Input/Output) operations or bytes written or read to each zone;

mapping the tracked number of reads and writes to a counter which tracks a physical region of the memory device that is impacted by the reads and writes;

determining a highest-to-lowest ranking of memory device wear based on the mapping; and

during compaction of the memory device, when a memory device log is re-written, moving the memory device log to a location with a lower ranking.

16. The method of claim 15, wherein the lower ranking is a lowest ranking.

17. The method of claim 15, wherein the tracked number of reads and writes is based on any of:

bytes, slices, blocks, or time spent reading and/or writing.