Caching techniques

• US 8,713,261 B1
• Filed: 03/11/2011
• Issued: 04/29/2014
• Est. Priority Date: 03/11/2011
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Aharoni, Dan, Naamad, Amnon, Veprinsky, Alex, Don, Arieh
• Primary Examiner(s): Dudek, Jr., Edward
• Assistant Examiner(s): BIRKHIMER, CHRISTOPHER D
• Application Number:
  US13/065,016
• Time in Patent Office: 1,145 Days
• Field of Search: None
• US Class Current: 711/137
• CPC Class Codes: G06F 12/0868 Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
• CPC Class Codes: G06F 2212/1024 Latency reduction
• CPC Class Codes: G06F 2212/286 Mirrored cache memory
• CPC Class Codes: G06F 2212/312 In storage controller

Abstract

Described are techniques for caching. At a first point in time, a first set of data portions currently stored in a first cache of a first data storage system is determined. Each data portion of the first set is located on a first device of the first data storage system. Sent to a second data storage system is first information including information identifying a storage location on the first device for each data portion of the first set. The second data storage system includes a second device that is maintained as a mirror of the first device. The storage location for each data portion of the first set is used to identify a second storage location of the second device corresponding to the storage location. The first information is used to populate a second cache of the second data storage system.

Claims

1. A method for caching comprising:

determining, at a point in time, a first set of data portions currently stored in a first cache of a first data storage system, each data portion of said first set being located on a first device of the first data storage system;

sending, to a second data storage system, first information including information identifying a storage location on said first device for each data portion of said first set, wherein said second data storage system includes a second device that is maintained as a mirror of the first device, said storage location for each data portion of said first set used to identify a second storage location of said second device corresponding to said storage location; and

using said first information to populate a second cache of the second data storage system.

2. The method of claim 1, further comprising:

determining one or more caching profiles, wherein each of said caching profiles identifies a plurality of data portions stored in cache and each of said caching profiles is associated with criteria identifying when to use said each caching, profile to pre-populate a cache, wherein at least a first of said one or more caching profiles is associated with criteria including any of date information and time information identifying, a time of day, wherein, for each of said plurality of data portions identified by said first caching profile as being stored in said cache, said first caching profile includes one or more indicators including an age indicator indicating any of when said each data portion was last used and how long said each data portion has been in cache;

selecting said first caching profile for use in pre-populating the cache prior to usage at a second point in time, wherein said first caching profile is associated with criteria matching said second point in time; and

pre-populating the cache prior to said second point in time with data portions identified in said first caching profile and with other information including said age indicator of the first caching profile for each data portion stored in the cache by said pre-populating.

3. The method of claim 2, wherein the method further includes:

processing a plurality of data operations directed to one or more devices of a data storage system;

recording what data portions are stored in cache at one or more points in time while processing said plurality of data operations; and

creating said one or more caching profiles, each of said caching profiles being associated with one of said points in time and identifying a plurality of data portions stored in cache in accordance with said recording.

4. The method of claim 2, wherein criteria associated with the first caching profile includes information characterizing activity causing the data portions to be stored in cache.

5. The method of claim 2, wherein criteria associated with the first caching profile identifies any of a day of a week, one or more days of a month, an application, and one or more months.

6. The method of claim 2, wherein the cache is included in a data storage system.

7. The method of claim 2, wherein each of the caching profiles identifies at least one data portion stored in cache by identifying a physical location on a data storage device associated with said at least one data portion.

8. The method of claim 7, wherein the physical location is identified using a plurality of coordinates.

9. The method of claim 8, wherein the plurality of coordinates includes a device identifier, cylinder, and head or track identifier.

10. The method of claim 7, wherein, for said each data portion identified as being stored, in cache by said each caching profile, said one or more indicators for said each data portion included in said each caching profile includes a data operation indicator regarding whether a read data operation or a write data operation has been most recently performed using said each data portion.

11. The method of claim 1, wherein the first data storage system fails and the second data storage system services data requests directed to the first data storage system, wherein data requests directed to the first device of the first data storage system are serviced using corresponding data portions from the second device and second cache of the second data storage system.

12. The method of claim 1, further comprising:

automatically propagating writes that are applied to the first device of the first data storage system to the second device of the second data storage system.

13. The method of claim 12, wherein a remote data facility is used to automatically propagate the writes that are applied to the first device from the first data storage system to the second data storage system and said remote data facility is used to send the first information to the second data storage system.

14. The method of claim 1, wherein a hash table is used to determine said first set of data portions.

15. The method of claim 14, wherein said hash table includes a plurality of entries identifying a plurality of cache locations of the first cache, each of said cache locations having stored therein a data portion of the first set.

16. The method of claim 1, wherein, for each data portion of the first set, the first information includes a timestamp associated with said each data portion, said timestamp indicating a cache age associated with said each data portion.

17. The method of claim 1, further comprising:

traversing a hash table to identify cache slots including cached data from the first device, and

using a mapping function to determine a storage device location on the first device corresponding to a hash entry of the hash table.

18. An apparatus comprising:

a first data storage system including a first cache and a first device;

a second data storage system including a second cache and a second device; and

a non-transitory computer readable medium of executable code stored thereon for;

determining, at a point in time, a first set of data portions currently stored in the first cache, each data portion of said first set being located on the first device;

sending, from the first data storage system to the second data storage system, first information including information identifying a storage location on said first device for each data portion of said first set, wherein said second device is maintained as a mirror of the first device, said storage location for each data portion of said first set used to identify a second storage location of said second device corresponding to said storage location; and

using said first information to populate the second cache of the second data storage system.

19. A non-transitory computer readable medium comprising code stored thereon for caching, the non-transitory computer readable medium comprising code for:

determining, at a point in time, a first set of data portions currently stored in a first cache of a first data storage system, each data portion of said first set being located on a first device of the first data storage system;

sending, to a second data storage system, first information including information identifying a storage location on said first device for each data portion of said first set, wherein said second data storage system includes a second device that is maintained as a mirror of the first device, said storage location for each data portion of said first set used to identify a second storage location of said second device corresponding to said storage location; and

using said first information to populate a second cache of the second data storage system.

20. The non-transitory computer readable medium of claim 19, wherein the first data storage system fails and the second data storage system services data requests directed to the first data storage system, wherein data requests directed to the first device of the first data storage system are serviced using corresponding data portions from the second device and second cache of the second data storage system.