PREDICTIVE PRE-LAUNCH FOR APPLICATIONS

• US 20140372356A1
• Filed: 06/12/2013
• Published: 12/18/2014
• Est. Priority Date: 06/12/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Iyigun, Mehmet, Kralj, Milos, Pulapaka, Hari, Kishan, Arun, Gunawardana, Asela, Meek, Christopher, Horvitz, Eric, Fortin, Michael, Bilal, Ahmad, Kleynhans, Christopher, Koch, Paul, Caruana, Rich
• Granted Patent: US 9,508,040 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06F 9/445 Program loading or initiating bootstrapping G06F9/4401; security arrangements for program loading or initiating G06F21/57
• CPC Class Codes: G06F 9/44578 Preparing or optimising for loading
• CPC Class Codes: G06F 9/485 Task life-cycle, e.g. stopping, restarting, resuming execution G06F9/4881 takes precedence
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

Systems and methods of pre-launching applications in a computer system, said applications being likely to be activated by a user from a terminated and/or suspended process state, are disclosed. The pre-launching of an application may be based on the assessed probability of the application being activated—as well as the level of availability of system resources to affect such pre-launching. Applications may be pre-launched based on these and other conditions/considerations, designed to improve the user'"'"'s experience of a quick launch of applications in the background. Several prediction models are presented to provide a good estimate of the likelihood of an application being activated by a user. Such prediction models may comprise an adaptive predictor (based on past application usage situations) and/or a switch rate predictor (based on historic data of an application being switched and, possibly, having a decay rate applied to such switch rate measure).

Claims

1. A method for pre-launching applications within a computer system said computer system comprising a controller, a memory and an operating system, the method comprising:

for a given set of applications, said applications capable of being activated by a user of the computer system, associating a prediction of when said applications may be activated by said user;

monitoring system resources of said computer system;

applying a set of pre-launch rules to said set of applications, said pre-launch rules comprising one of a group, said group comprising;

rules regarding availability of said system resources and rules regarding said predictions associated with said applications; and

pre-launching one said application, based upon the satisfaction of said set of pre-launch rules.

2. The method of claim 1 wherein said applications capable of being activated by a user comprise one of a group, said group comprising:

individual applications, applications that are used together by said user, applications that are used in a certain sequence, applications that are used at certain times of the day, applications that are used at regular intervals, applications used at a certain location and applications that are used in response to certain events in the computer system.

3. The method of claim 1 wherein said associating a prediction of when said application may be activated further comprises:

prediction modeling to give a prediction measure of when an application may be activated by a user; and

further wherein said prediction modeling is one of a group, said group comprising;

modeling based on order of application usage, modeling based on frequency of application usage, modeling based on time of day of application usage, modeling based on location of application usage, modeling using most common application predictor, modeling using most used predictor, modeling using a null predictor, modeling using an oracle predictor, modeling with an adaptive predictor and modeling with a switch rate predictor.

4. The method of claim 3 wherein said modeling with an adaptive predictor further comprises:

identifying past application usage situations;

comparing the current application usage situation;

returning a measure that a queried application may be activated within a desired prediction window.

5. The method of claim 4 wherein said situations may comprise one of a group, said group comprising:

the current foreground application, the last foreground application and how long the current application has been in usage.

6. The method of claim 3 wherein said modeling with a switch rate predictor further comprises:

maintaining switch rate data on an application over time; and

providing a measure for when said application may be likely to be switched.

7. The method of claim 6 wherein said modeling with a switch rate predictor further comprises:

applying a decay rate to said switch data rate over time; and

changing said measure according to said decay rate.

8. The method of claim 3 wherein said monitoring system resources further comprises:

monitoring a level of availability of said system resources to said computer system; and

wherein further said system resources comprises one of a group, said group comprising;

CPU utilization, GPU utilization, memory utilization, battery state of charge and I/O utilization.

9. The method of claim 8 wherein said applying a set of pre-launch rules further comprises:

considering said level of availability of said system resources to a desired threshold;

considering the prediction measure of when an application may be activated by a user; and

determining whether to pre-launch said application depending upon the satisfaction of conditions based on said level of availability and said prediction measure.

10. The method of claim 1 wherein each of said applications capable of being activated by a user comprises an associated process state, said associated process state comprises one of a group, said group comprising:

running, not running and suspended.

11. The method of claim 10 wherein pre-launching one said application further comprises:

changing the associated process state of one said application from not running to running.

12. The method of claim 11 wherein changing the associated process state of one said application further comprises:

changing the associated process state of one said application in the background of said computer system.

13. The method of claim 12 wherein changing the associated process state of one said application further comprises:

activating said one said application to its last saved state with its backstack location restored.

14. The method of claim 1 wherein said method further comprises:

blacklisting a given application that has failed pre-launch in the past.

15. The method of claim 1 wherein said method further comprises:

seeding a given computer system with initial prediction data from one of a group, said group comprising;

data collected a community of users and data from different computer systems.

16. A system for pre-launching applications on a computer system, said computer system comprising a controller, a memory and an operating system, the system comprising:

a prediction engine, said prediction engine capable of providing a prediction measure of likelihood that a given application may be activated by a user of said computer system;

a resource monitor, said resource monitor capable of providing a measure of system resource utilization of said computer system;

a pre-launch policy module, said pre-launch policy module capable of applying a set of pre-launch policy rules, said pre-launch policy rules comprising one of a group, said group comprising;

rules regarding availability of said system resources and rules regarding said prediction measures associated with said applications; and

a pre-launch queue module, said pre-launch queue module capable of maintaining a list of applications for pre-launching in said computer system, said list of applications depending upon the satisfaction of said pre-launch policy rules.

17. The system of claim 16 wherein said prediction engine further comprises:

at least one prediction module, said prediction module comprising one of a group, said group comprising;

prediction module based on order of application usage, prediction module based on frequency of application usage, prediction module based on time of day of application usage, prediction module based on location of application usage, prediction module using most common application predictor, prediction module using most used predictor, prediction module using a null predictor, prediction module using an oracle predictor, prediction module with an adaptive predictor and prediction module with a switch rate predictor.

18. The system of claim 17 wherein said system is capable of blacklisting a given application that has failed pre-launch in the past.

19. A computer-readable storage media storing instructions that when executed by a computing device, said instructions cause the computing device to perform operations comprising:

for a given set of applications, said applications capable of being activated by a user of the computer system, associating a prediction of when said applications may be activated by said user with said application;

monitoring system resources of said computer system;

applying a set of pre-launch rules to said set of applications, said pre-launch rules comprising one of a group, said group comprising;

rules regarding availability of said system resources and rules regarding said predictions associated with said applications; and

pre-launching one said application, based upon the satisfaction of said set of pre-launch rules.

20. The computer-readable storage medium of claim 19 wherein said associating a prediction of when said application may be activated further comprises:

prediction modeling to give a prediction measure of when an application may be activated by a user; and

further wherein said prediction modeling is one of a group, said group comprising;

modeling based on order of application usage, modeling based on frequency of application usage, modeling based on time of day of application usage, modeling based on location of application usage, modeling using most common application predictor, modeling using most used predictor, modeling using a null predictor, modeling using an oracle predictor, modeling with an adaptive predictor and modeling with a switch rate predictor.