PREDICTIVE CACHING AND FETCH PRIORITY

• US 20150193347A1
• Filed: 01/06/2014
• Published: 07/09/2015
• Est. Priority Date: 01/06/2014
• Status: Active Grant

• Current Assignee: Dropbox, Inc.
• Original Assignee: Dropbox, Inc.
• Inventors: Kluesing, Daniel, Andersson, Rasmus
• Granted Patent: US 9,411,814 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 12/0862 with prefetch
• CPC Class Codes: G06F 12/0888 using selective caching, e.g. bypass
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 2212/602 Details relating to cache prefetching
• CPC Class Codes: G06F 2212/6026 Prefetching based on access pattern detection, e.g. stride based prefetch
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence

Abstract

Predicting what content items a user finds important and sending those items to a cache on the user'"'"'s device at times when doing so will not drain resources and will not result in expensive data rates. Applying a ranking function that examines recency and other content metadata associated with the user'"'"'s content items stored in a synchronized content management system. Determining how much of a ranked list of content items to cache and deciding when is a good time to send content items to the local cache.

Claims

1. A computer-implemented method comprising:

receiving, from a mobile device, an amount of free memory available for caching one or more content items from a collection of content items stored in a synchronized content management system, wherein each of the content items in the collection of content items has associated metadata;

generating, via at least one processor, a scaling factor for each of the content items in the collection based on the associated metadata; and

selecting a subset of content items from the collection based on the generated scaling factors and the amount of free space.

2. The computer-implemented method of claim 1, wherein selecting a subset of content items from the collection comprises:

generating a ranked list of content items from the collection, wherein an item with a higher scaling factor has a higher rank in the ranked list; and

selecting content items from the ranked list beginning with higher ranked content items until an amount of space required to cache the selected content items fills the amount of free space.

3. The method of claim 2 further comprising:

selecting a plurality of representation quality levels for a content item from the ranked list.

4. The computer-implemented method of claim 2 further comprising:

automatically caching the selected content items to the mobile device.

5. The method of claim 4, further comprising conducting a mobile device stamina check before caching the selected content items to the mobile device, wherein the stamina check comprises:

determining whether the mobile device has a battery level above a predetermined minimum battery level before automatically caching the selected content items;

determining whether a network connection for the mobile device satisfies a predetermined network connectivity level before automatically caching the selected content items; and

determining whether caching a content item from a synchronized content management system over a current network connection will result in a cellular network data fee before automatically caching a content item.

6. The method of claim 1, wherein the amount of free space is expressed as a predetermined percentage of the total free space in the memory of the mobile device.

7. The method of claim 1, wherein metadata associated with a content item comprises:

a FileScore calculated based on one or more recency-related events selected from among;

a most recent occurrence of a user of the mobile device opening the content item using an application offered by the synchronized content management system,

a most recent occurrence of the user of the mobile device opening the content item using a third party application,

a most recent occurrence of the user of the mobile device sharing the content item via the mobile device,

a most recent occurrence of another user of the synchronized content management system sharing the content item with the user of the mobile device,

a most recent occurrence of the user of the mobile device sharing the content item via another electronic device,

a most recent occurrence of the user of the mobile device editing the content item, and

a most recent occurrence of the user of the mobile device accessing the content item via another electronic device.

8. The method of claim 7, further comprising applying a weighting factor to the one or more recency-related events.

9. The method of claim 1, wherein metadata associated with a content item comprises:

a PinnedImportance variable for each content item in the collection of content items, wherein the binary PinnedImportance variable indicates a user specified importance level.

10. The method of claim 1, wherein metadata associated a content item comprises:

a SyncScore indicating a content item modification frequency.

11. The method of claim 1, wherein metadata associated a content item comprises:

an ObjectiveImportance value indicating a degree to which the synchronized content management system has deemed the content item as important.

12. A system comprising:

a processor;

an account management module configured to receive, from a mobile device, an amount of free memory available for caching one or more content items from a collection of content items stored in a user account database of a synchronized content management system, wherein each of the content items in the collection of content items has associated metadata; and

a cache prioritization module configured to;

generate a scaling factor for each of the content items in the collection based on the associated metadata;

generate a ranked list of content items from the collection, wherein an item with a higher scaling factor has a higher rank in the ranked list;

select a subset of content items from the collection based on the generated scaling factors and the amount of free space, such that higher ranked content items are selected until an amount of space required to cache the selected content items fills the amount of free space; and

automatically cache the selected content items to the mobile device.

13. The computer-implemented method of claim 12, wherein metadata associated with a content item comprises at least one of a FileScore value, a PinnedImportance value, a SyncScore value, or an ObjectiveImportance value.

14. The computer-implemented method of claim 13, wherein the FileScore value is calculated based on one or more recency-related events selected from among:

a most recent occurrence of a user of the mobile device opening the content item using an application offered by the synchronized content management system,

a most recent occurrence of the user of the mobile device opening the content item using a third party application,

a most recent occurrence of the user of the mobile device sharing the content item via the mobile device,

a most recent occurrence of another user of the synchronized content management system sharing the content item with the user of the mobile device,

a most recent occurrence of the user of the mobile device sharing the content item via another electronic device,

a most recent occurrence of the user of the mobile device editing the content item, and

a most recent occurrence of the user of the mobile device accessing the content item via another electronic device.

15. The computer-implemented method of claim 13, wherein the PinnedImportance value comprises a variable that indicates a user specified importance level.

16. The computer-implemented method of claim 13, wherein the SyncScore value comprises a variable indicating a content item modification frequency.

17. The computer-implemented method of claim 13, wherein the ObjectiveImportance value comprises a variable indicating a degree to which the synchronized content management system has deemed the content item as important.

18. A non-transitory computer-readable medium comprising:

a medium configured to store computer-readable instructions thereon; and

the computer-readable instructions that, when executed by a processing device cause the processing device to perform a method, comprising;

receiving, from a mobile device, an amount of free memory available for caching one or more content items from a collection of content items stored in a synchronized content management system, wherein each of the content items in the collection of content items has associated metadata;

generating a scaling factor for each of the content items in the collection based on the associated metadata;

generating a ranked list of content items from the collection, wherein an item with a higher scaling factor has a higher rank in the ranked list;

selecting a subset of the content items from the ranked list beginning with higher ranked content items until an amount of space required to cache the selected content items fills the amount of free space; and

automatically caching the selected content items to the mobile device.

19. The non-transitory computer-readable storage medium of claim 18, the instructions further causing the processing device to perform the step of conducting a mobile device stamina check before caching the selected content items to the mobile device, wherein the stamina check comprises:

determining whether the mobile device has a battery level above a predetermined minimum battery level before automatically caching the selected content items;

determining whether a network connection for the mobile device satisfies a predetermined network connectivity level before automatically caching the selected content items; and

determining whether caching a content item from a synchronized content management system over a current network connection will result in a cellular network data fee before automatically caching a content item.

20. The non-transitory computer-readable storage medium of claim 18, wherein metadata associated with a content item comprises a FileScore value that is calculated based on one or more recency-related events selected from among:

a most recent occurrence of a user of the mobile device opening the content item using an application offered by the synchronized content management system,

a most recent occurrence of the user of the mobile device opening the content item using a third party application,

a most recent occurrence of the user of the mobile device sharing the content item via the mobile device,

a most recent occurrence of another user of the synchronized content management system sharing the content item with the user of the mobile device,

a most recent occurrence of the user of the mobile device sharing the content item via another electronic device,

a most recent occurrence of the user of the mobile device editing the content item, and

a most recent occurrence of the user of the mobile device accessing the content item via another electronic device.