Distributed media cache

• US 10,223,275 B2
• Filed: 01/28/2016
• Issued: 03/05/2019
• Est. Priority Date: 01/28/2016
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Inventors: Feist, Jason M., Gaertner, Mark Allen, Purani, Dipeshkumar J., Kashyap, Anil, Zhao, Wei
• Primary Examiner(s): Yi, David
• Assistant Examiner(s): Ahmed, Zubair
• Application Number:
  US15/008,736
• Publication Number:
  US 20170220473A1
• Time in Patent Office: 1,132 Days
• CPC Class Codes: G06F 12/0871 Allocation or management of cache space
• CPC Class Codes: G06F 12/0873 Mapping of cache memory to specific storage devices or parts thereof
• CPC Class Codes: G06F 2212/1036 Life time enhancement

Abstract

Zones of a magnetic recording medium are allocated as a respective plurality of distributed media caches arranged in in a predetermined order. For each of a plurality of caching events, cache data is written to one or more of the distributed media caches. A next of the media caches in the predetermined order is selected for the next caching event if the selected caches are not a last in the predetermined order. Otherwise a first media cache is selected in the predetermined order.

Claims

1. A method comprising:

defining a first set of two or more adjacent tracks and a second set of two or more adjacent tracks on a surface of a disk as media cache zones, the first and second sets located at radial locations between a middle diameter and an outer diameter of the disk that are tolerant to a high level of write activity and that minimize wear of the media cache zones due to one or more of head-to-disk contact and thermal degradation, the first and second sets being separated by user data zones;

allocating the media cache zones as a plurality of distributed media caches arranged in a predetermined order within the media cache zones;

for each of a plurality of caching events;

writing cache data to a selected one or more of the distributed media caches; and

in response to the cache data being, written to the selected media cache for a predetermined time period, selecting a next of the distributed media caches in the predetermined order for the next caching event if the selected caches are not a last in the predetermined order, otherwise selecting a first media cache in the predetermined order if the selected distributed media cache is the last of the predetermined order.

2. The method of claim 1, wherein the first set of adjacent tracks further comprises first additional adjacent tracks on a second surface of the disk opposed to the surface, the first additional adjacent tracks vertically aligned with the first set of adjacent tracks, and wherein the second set of adjacent tracks further comprises second additional adjacent tracks on the second surface of the disk, the second additional adjacent tracks vertically aligned with the second set of adjacent tracks.

3. The method of claim 1, wherein the media cache zones comprise additional sets of adjacent tracks, each set being between additional user data zones.

4. The method of claim 1, wherein the caching event comprise copying data from a dynamic, random access memory to the media caches.

5. The method of claim 1, wherein the caching event comprise copying data from a user data sector to the media caches in response to a host read request.

6. The method of claim 1, further comprising:

reducing a storage capacity of one of the media cache zones; and

in response the reduction of the storage capacity, utilizing an overprovisioned storage space of others of the media cache zones to compensate for the reduction of the storage capacity.

7. The method of claim 6, wherein the overprovisioned storage space comprises guard bands between the media cache zones and the user data zones, and wherein utilizing the overprovisioned storage space comprises mapping one or more tracks of the guard bands to logical addresses.

8. The method of claim 6, wherein the overprovisioned storage space is configured with a linear bit density that is less than a maximum, and wherein utilizing the overprovisioned storage space comprises increasing the linear bit density.

9. The method of claim 1, wherein the predetermined order is a random order.

10. The method of claim 1, wherein the predetermined time is based on a write time to the selected media cache.

11. The method of claim 1, wherein the predetermined time is based on a system time that ensures no repeated access to the selected media cache for the predetermined time.

12. The method of claim 1, wherein the predetermined time is based on an amount of data written to the predetermined media cache.

13. The method of claim 1, wherein allocating the media cache zones comprises overprovisioning the media cache zones to prevent undue wear in the media cache regions as the result of a field re-manufacturing operation.

14. An apparatus comprising:

interface circuitry coupled to a read/write head configured to access a surface of a disk, a first set of two or more adjacent tracks and a second set of two or more adjacent tracks on the surface of the disk being defined as media cache zones, the first and second sets located at radial locations between a middle diameter and an outer diameter of the disk that are tolerant to a high level of write activity and that minimize wear of the media cache zones due to one or more of head-to-disk contact and thermal degradation, the first and second sets being separated by user data zones, the media cache zones being allocated as a plurality of distributed media caches arranged in a predetermined order within the media cache zones;

a controller coupled to the interface circuitry and configured to perform for each of a plurality of caching events;

writing cache data to a selected one or more of the distributed media caches; and

in response to the cache data being written to the selected media cache for a predetermined time period, selecting a next of the distributed media caches in the predetermined order for the next caching event if the selected caches are not a last in the predetermined order, otherwise selecting a first media cache in the predetermined order if the selected distributed media cache is the last of the predetermined order.

15. The apparatus of claim 14, wherein the first set of adjacent tracks further comprises first additional adjacent tracks on a second surface of the disk opposed to the surface, the first additional adjacent tracks vertically aligned with the first set of adjacent tracks, and wherein the second set of adjacent tracks further comprises second additional adjacent tracks on the second surface of the disk, the second additional adjacent tracks vertically aligned with the second set of adjacent tracks.

16. The apparatus of claim 14, wherein the predetermined time is based on at least one of a write time to the selected media cache, a system time that ensures no repeated access to the selected media cache for the predetermined time, and an amount of data written to the predetermined media cache.

17. The apparatus of claim 14, wherein allocating the media cache zones comprises overprovisioning the media cache zones to prevent undue wear in the media cache regions as the result of a field re-manufacturing operation.

18. A method comprising:

defining a first set of two or more adjacent tracks and a second set of two or more adjacent tracks on a surface of a disk as media cache zones, the first and second sets located between a middle diameter and an outer diameter of the disk and being separated by user data zones;

allocating the media cache zones as a plurality of distributed media caches arranged in a predetermined order within the media cache zones, a capacity of the media cache zones being overprovisioned to prevent undue wear in the media cache zones, wherein the overprovisioned capacity comprises one of;

guard bands between the media cache zones and adjacent data regions; and

a low linear bit density for the media cache zones; and

for each of a plurality of caching events;

writing cache data to a selected one or more of the distributed media caches; and

in response to the cache data being written to the selected media cache for a predetermined time period, selecting a next of the distributed media caches in the predetermined order for the next caching event if the selected caches are not a last in the predetermined order, otherwise selecting a first media cache in the predetermined order if the selected distributed media cache is the last of the predetermined order.

19. The method of claim 18, further comprising recovering the overprovisioned capacity of the media cache zones by adjusting the guard bands between the media cache zones and adjacent data regions.

20. The method of claim 18, further comprising recovering the overprovisioned capacity of the media cache zones by increasing a linear bit density used within the media cache zones.