Utility-based archiving

• US 7,548,904 B1
• Filed: 11/23/2005
• Issued: 06/16/2009
• Est. Priority Date: 06/28/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Horvitz, Eric J.
• Primary Examiner(s): Rones; Charles
• Assistant Examiner(s): Choi; Yuk Ting
• Application Number:
  US11/286,667
• Time in Patent Office: 1,301 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

A system and associated methodology is provided that is adapted to infer what to do with an item, and more particularly whether to archive and/or keep active an item in a more active, easy-to-access store based upon a cost-benefit analysis. The cost-benefit analysis determines the overhead associated with keeping the item active (e.g., not archiving it) versus the gains in connection with having quick and easy access to the item. The cost of maintaining an item in an active state is measured in terms of the size of the item which, in turn, affects the amount of space needed to store it. The benefit of keeping the item active is measured in terms of a probabilistic determination describing how a user will access the item in the future, which is a reflection of the utility of the item in an accessible state. The invention leverages notions of temporal sensitivity of the likelihood that an item will be needed in the future such that determined values and inferences can be dynamically updated over time. Items having a small probability of being accessed again after an initial review are categorized as one-shot items.

Claims

1. A system that facilitates maintaining an item, comprising:

an inference component that inferentially determines whether to store the item in an active or archived state based at least in part upon;

a probabilistic cost-benefit analysis to determine the item'"'"'s worth by comparing an overhead cost attributable to storing the item in the active state versus a benefit associated with storing the item in the active state, the overhead cost determined based in part on the size of the item and the benefit determined at least in part by calculating a probability a user will access the item in the future;

an assessment of the item'"'"'s worth relative to the worth of other items actively stored, wherein the item is stored in the active or archived state according to the assessment of the item'"'"'s relative worth;

an assessment of the number of times the item is likely to be accessed by the user, wherein an item that is unlikely to be viewed more than once is stored in the active state until viewed by the user; and

at least one of an archiving rule or a policy that affects the probabilistic cost-benefit analysis, the at least one archiving rule or policy is customized by a user; and

an optimization component that determines an arrangement of items in the active state to maximize a total value density of the items in the active state, the value density for the item is determined based upon determining the probability a user will access the item divided by a total size of the item,

wherein a memory operatively coupled to a processor retains at least one of the inference component or the optimization component.

2. The system of claim 1 further comprising:

a property log that stores user preferences gathered from implicit evidence obtained by monitoring user activity, the inference system consulting the property log when making an inferential determination.

3. The system of claim 2 being temporally sensitive such that information stored in the property log is updated over time.

4. The system of claim 1, the inference system utilizing a value density of the item to perform the cost-benefit analysis.

5. The system of claim 1 further comprising:

a learning system that acts upon the inference system to modify inferences based upon at least one of;

a property of the item and a property of the user.

6. The system of claim 5 wherein the inference system employs a knapsack packing analysis to optimize items stored in the active state.

7. The system of claim 1 being temporally sensitive such that the item'"'"'s worth is updated over time.

8. The system of claim 1 further comprising an interactive user interface.

9. The system of claim 8 wherein the interactive user interface comprises a selection element to allow a condition to be enabled/disabled.

10. The system of claim 9 wherein the interactive user interface further comprises an entry element to allow a condition to be configured.

11. The system of claim 1, wherein the system operates in a networking environment.

12. A computer implemented method for determining how to store a plurality of items comprising:

determining respective item utilities through probabilistic computations;

determining respective item value densities in accordance with a probability of future item access and item size, the value density is the probability of user access of the item divided by the total size of the item;

assessing a worth of one or more items relative to at least one other of the plurality of items utilizing the item utilities and item value densities;

inferring whether to remove one or more of the plurality of items from active memory based upon the one or more items'"'"' worth relative to at least one other of the plurality of items;

determining which items to retain in the active memory when the active memory is full based on a comparison of the value densities of two or more items in the active memory;

removing one or more items from active memory according to the inference;

classifying an item stored in active memory as a one-shot item based on the number of times the item is likely to be accessed by a user,

removing the one-shot item from the active memory once the one-shot item is viewed by the user; and

enabling a user to personalize how items are stored by customizing at least one of an archiving rule or a policy that affects the probabilistic computations.

13. The computer implemented method of claim 12 further comprising:

updating item utilities over time.

14. The computer implemented method of claim 13 further comprising:

determining modern value densities based upon updated item utilities.

15. The computer implemented method of claim 14 further comprising:

making contemporary inferences from at least one of;

updated utilities and modern value densities.

16. The computer implemented method of claim 15 further comprising:

mitigating inefficient use of active space by applying a knapsack computation to the modern value densities.

17. The computer implemented method of claim 12 further comprising:

learning how to adjust inferences based upon at least one of;

updated log information, probabilities and value densities.

18. A utility based item archiving system comprising:

means for determining a utility of an item, the utility determined at least in part by calculating a probability a user would prefer the item stored in a fast memory;

means for determining one or more costs associated with storing the item in the fast memory, at least one cost relating to the size of the item;

means for determining a relative worth of the item, the means for determining a relative worth comparing the utility and one or more costs of the item to that of at least one other item;

means for inferring whether to store the item in the fast memory or in an archival memory based upon the relative worth of the item;

means for classifying an item stored in the fast memory as a one-shot item based on tile number of times the item is likely to be accessed by a user, and

means for removing the one-shot item from the fast memory once the one-shot item is viewed by the user; and

means for enabling a user to customize at least one of an archiving rule or a policy that affects the probability calculation;

means for determining an arrangement of items in the fast memory to maximize a total value density of the items in the fast memory, the value density for the item is determined based upon determining the probability a user will access the item divided by a total size of the item, wherein a memory operatively coupled to a processor retains at least one of the means.

19. The system of claim 18, wherein the system operates in a networking environment.