Automated device-specific dynamic operation modifications

• US 10,248,527 B1
• Filed: 09/19/2018
• Issued: 04/02/2019
• Est. Priority Date: 09/19/2018
• Status: Active Grant

• Current Assignee: Curinos Incorporated
• Original Assignee: Amplero Inc.
• Inventors: Miller, Scott Allen, Hersch, Jesse, Cazzanti, Luca, Downs, Oliver B.
• Primary Examiner(s): Dean, Jr., Joseph E
• Application Number:
  US16/136,035
• Time in Patent Office: 195 Days
• CPC Class Codes: G06F 1/3206 Monitoring of events, devices or parameters that trigger a change in power modality
• CPC Class Codes: G06F 1/3234 Power saving characterised by the action undertaken
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3051 Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs verification or detection of system hardware configuration G06F11/2247
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 9/44505 Configuring for program initiating, e.g. using registry, configuration files
• CPC Class Codes: G06F 9/4893 taking into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: H04M 1/72448 with means for adapting the functionality of the device according to specific conditions
• CPC Class Codes: H04W 52/0258 controlling an operation mode according to history or models of usage information, e.g. activity schedule or time of day
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management
• CPC Class Codes: Y02D 30/70 in wireless communication networks

Abstract

Techniques are described for automatically and dynamically modifying ongoing operation of computing devices in device-specific manners, such as to improve ongoing performance of the computing devices by modifying configuration settings on the computing devices or on associated systems in communication with the computing devices. The techniques may include generating one or more decision structures that are each specific to a type of measured performance effect, and using the decision structure(s) to improve corresponding performance of a computing device, with the generating of the decision structure(s) including analyzing training data that associates prior measured performance effects with corresponding attributes of computing devices and of modification actions that were performed for the computing devices. Non-exclusive examples of modifying operation of a smartphone computing device include modifying configuration settings affecting use of one or more of the device'"'"'s battery, memory, storage and network communications.

Claims

1. A computer-implemented method comprising:

tracking, by a configured computing system, device operations of a plurality of mobile devices, including generating data about modification actions of multiple types that are performed on configuration settings affecting use of hardware components on the mobile devices, and about subsequent changes in the device operations of the mobile devices;

determining, by the configured computing system and based on an automated analysis of the generated data, measured effects on the device operations that result from performing associated modification actions on the configuration settings, wherein the mobile devices have a plurality of device attributes reflecting the hardware components on the mobile devices, and wherein the associated modification actions of the multiple types have a plurality of action attributes reflecting associated configuration setting modifications;

generating, by the configured computing system, a decision tree structure for a subset of the device and action attributes that are selected based on having associated measured effects of an indicated type, including generating a node in the decision tree structure associated with each of multiple combinations of the device and action attributes of the subset, and storing a target distribution and a control distribution for each node to identify measured effects for devices having the device attributes of the associated combination for the node, wherein the target distribution for a node identifies the measured effects for devices receiving modification actions with the action attributes of the associated combination for the node, and wherein the control distribution for a node identifies the measured effects for other devices that do not receive modification actions with the action attributes of the associated combination for the node; and

using, by the configured computing system, the decision tree structure to control ongoing operations for an additional mobile device, including;

determining, by the configured computing system, and for each of multiple modification actions to possibly perform, a node in the decision data structure with an associated combination of device and action attributes that matches attributes of the additional mobile device and the modification action, and using the target and control distributions for the determined node to predict an effect of performing the modification action on the additional mobile device;

selecting, by the configured computing system, one of the multiple modification actions based at least in part on the predicted effect of performing the selected one modification action on the additional mobile device; and

performing, by the configured computing system, the selected one modification action on the additional mobile device to modify one or more configuration settings that affect use of hardware components on the additional mobile device.

2. The computer-implemented method of claim 1 further comprising:

measuring, by the configured computing system and after the performing of the selected one modification action on the additional mobile device, actual performance effects on the additional mobile device associated with the selected one modification action; and

updating, by the configured computing system, information in the decision tree structure to correspond at least in part to the actual performance effects.

3. The computer-implemented method of claim 1 wherein the using of the target and control distributions for a determined node to predict an effect of performing a modification action on the additional mobile device includes performing a weighted sampling of the measured effects identified in the target distribution to determine a first prediction corresponding to modification action performance, performing a weighted sampling of the measured effects identified in the control distribution to determine a second prediction corresponding to modification action non-performance, and determining the predicted effect of performing the modification action on the additional mobile device as a difference between the first and second predictions.

4. A computer-implemented method comprising:

obtaining, by a configured computing system, a decision data structure generated from data about previous modification actions of multiple types that are performed on configuration settings affecting use of hardware components on mobile devices and that are associated with subsequent changes in device operations of the mobile devices, wherein the mobile devices have a plurality of device attributes and the modification actions have a plurality of action attributes, and wherein the decision data structure includes information about measured effects for a subset of the device and action attributes that are selected based on combinations of those device and action attributes being associated with measured effects of an indicated type in the subsequent changes; and

using, by the configured computing system, the decision data structure to control ongoing operations for an additional mobile device, including;

determining, by the configured computing system, and for each of multiple modification actions to possibly perform, one of the combinations of device and action attributes in the decision data structure that matches attributes of the additional mobile device and the modification action, and using information about the measured effects for the determined combination to predict an effect of performing the modification action on the additional mobile device;

selecting, by the configured computing system, one of the multiple modification actions to perform based at least in part on the predicted effect of performing the selected one modification action on the additional mobile device; and

initiating, by the configured computing system, performance of the selected one modification action on the additional mobile device to modify one or more configuration settings that affect use of hardware components on the additional mobile device.

5. The computer-implemented method of claim 4 wherein the measured effects for the subset of the device and action attributes are based on performance of batteries powering the mobile devices, wherein the predicted effect of performing the selected one modification action on the additional mobile device involves increasing a length of time that one or more batteries are available to power the additional mobile device, and wherein the initiating of the performance of the selected one modification action on the additional mobile device includes reducing use of one or more hardware components on the additional mobile device that are associated with battery usage.

6. The computer-implemented method of claim 5 wherein the reducing of the use of one or more hardware components on the additional mobile device includes modifying one or more configuration setting on the additional mobile device to reduce use of a display on the additional mobile device.

7. The computer-implemented method of claim 5 wherein the reducing of the use of one or more hardware components on the additional mobile device includes modifying one or more configuration setting on one or more other devices that are associated with managing electronic communications sent to and/or from the additional mobile device to reduce use of a network interface on the additional mobile device.

8. The computer-implemented method of claim 4 wherein the measured effects for the subset of the device and action attributes are based on errors in network communications involving the mobile devices, wherein the predicted effect of performing the selected one modification action on the additional mobile device involves reducing errors of one or more types for network communications involving the additional mobile device, and wherein the initiating of the performance of the selected one modification action on the additional mobile device includes increasing a memory buffer size on the additional mobile device that is associated with streaming data to the additional mobile device.

9. The computer-implemented method of claim 4 wherein the measured effects for the subset of the device and action attributes are based on at least one of processor or memory performance on the mobile devices, wherein the predicted effect of performing the selected one modification action on the additional mobile device involves improving the at least one of processor or memory performance on the additional mobile device, and wherein the initiating of the performance of the selected one modification action on the additional mobile device includes reducing use of virtualization on the additional mobile device.

10. The computer-implemented method of claim 4 wherein the configured computing system is a server system separated from the additional mobile device by one or more computer networks, and wherein the using of the decision data structure to control ongoing operations for an additional mobile device is performed for each of a plurality of additional mobile devices.

11. The computer-implemented method of claim 4 wherein the configured computing system is the additional mobile device, and wherein the using of the decision data structure includes executing an automated performance manager system on the additional mobile device to perform the controlling of the ongoing operations for the additional mobile device.

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

obtaining additional decision data structures that each includes information about a type of measured effect for a subset of device and action attributes that are selected for use in that decision data structure, and wherein the type of measured effect for each of the additional decision data structures is different from that for the obtained decision data structure and from other of the additional decision data structures; and

determining, by the configured computing system, and for each of the additional decision data structures, a predicted effect of the type for the additional decision data structure for performing each of the multiple modification actions on the additional mobile device;

and wherein the selecting of the one modification action further includes blending the determined predicted effects from each of the additional decision data structures for the one modification action with the predicted effect from the decision data structure for the one modification action.

13. The computer-implemented method of claim 12 wherein the determining of the predicted effect for each of the additional decision data structures for performing each of the multiple modification actions on the additional mobile device includes identifying one of multiple nodes in the additional decision data structure that is associated with a combination of device and action attributes in the decision data structure that matches attributes of the additional mobile device and that modification action, and wherein the blending of the determined predicted effects from each of the additional decision data structures for the one modification action with the predicted effect from the decision data structure for the one modification action includes weighting the determined predicted effect from each additional decision data structure based at least in part on how similar the device and modification attributes for the identified node in that additional decision data structure for the one modification action are to an intersection of device and modification attributes of all nodes.

14. The computer-implemented method of claim 12 wherein the blending of the determined predicted effects from each of the additional decision data structures for the one modification action with the predicted effect from the decision data structure for the one modification action includes weighting the determined predicted effect from each additional decision data structure to minimize a mean square lift over all of the multiple modification actions for training data used to train the additional decision data structures.

15. The computer-implemented method of claim 4 wherein the using of the decision data structure to control ongoing operations for the additional mobile device further includes receiving, by the configured computing system, a request for a modification action to be performed substantially immediately, and wherein the determining and the using of the information and the selecting and the initiating are performed in response to the received request.

16. The computer-implemented method of claim 4 further comprising notifying, by at least one of the configured computing system or the additional mobile device, and after the performance of the selected one modification action on the additional mobile device, a user of the additional mobile device of the performance of the selected one modification action on the additional mobile device.

17. The computer-implemented method of claim 4 further comprising notifying, by at least one of the configured computing system or the additional mobile device, a user of the additional mobile device of the selected one modification action, and wherein the performance of the selected one modification action on the additional mobile device is further based at least in part on a response of the user to the notifying.

18. The computer-implemented method of claim 4 wherein the selecting of the one modification action is further based in part on additional information specific to the additional mobile device to customize the selecting to the additional mobile device.

19. The computer-implemented method of claim 4 wherein the selecting of the one modification action is further based in part on additional information specific to a user of the additional mobile device to customize the selecting to the user.

20. The computer-implemented method of claim 4 wherein some of the combinations of device and action attributes with information included in the decision data structure have action attributes that further correspond to actions involving modifying content being provided via programs executing on the mobile devices, wherein one or more of the multiple modification actions include modifying content being provided by at least one program executing on the additional mobile device, and wherein the method further comprises selecting and performing at least one of the one or more modification actions to cause content being provided by at least one program executing on the additional mobile device to be modified.

21. The computer-implemented method of claim 4 wherein some of the combinations of device and action attributes with information included in the decision data structure have action attributes that further correspond to actions involving providing choices to increase use of a service by the mobile devices, wherein one or more of the multiple modification actions include providing one or more choices to increase use of the service by the additional mobile device, and wherein the method further comprises selecting and performing at least one of the one or more modification actions to cause the providing of the one or more choices to increase use of the service by the additional mobile device.

22. The computer-implemented method of claim 4 wherein some of the combinations of device and action attributes with information included in the decision data structure have action attributes that further correspond to actions involving providing choices to retain use of a service by the mobile devices, wherein one or more of the multiple modification actions include providing one or more choices to retain use of the service by the additional mobile device, and wherein the method further comprises selecting and performing at least one of the one or more modification actions to cause the providing of the one or more choices to retain use of the service by the additional mobile device.

23. The computer-implemented method of claim 4 wherein the additional mobile device is at least one of a smartphone, a portable gaming device, a tablet computer or a laptop computer.

24. The computer-implemented method of claim 4 wherein the determined one combination of device and action attributes has an associated node in the decision data structure that has a target distribution and a control distribution, wherein the target distribution identifies the measured effects for devices with the device attributes of the determined one combination that receive modification actions with the action attributes of the determined one combination, wherein the control distribution identifies the measured effects for other devices with the device attributes of the determined one combination that do not receive modification actions with the action attributes of the determined one combination, and wherein predicting the effect of performing the modification action on the additional mobile device includes determining a difference in identified measured effects for the target distribution and identified measured effects for the control distribution.

25. The computer-implemented method of claim 4 wherein a quantity of all combinations of the plurality of device attributes and the plurality of action attributes is too large for the decision data structure to include information about measured effects for each of the all combinations, and wherein generating of the decision data structure further includes selecting the device and action attributes of the subset based at least in part on the selected device and action attributes providing an information gain related to correlations with the measured effects that is greater than information gains for other possible device and action attributes that are not selected.

26. A non-transitory computer-readable medium having stored contents that cause one or more computing systems to perform automated operations including at least:

obtaining, by a configured computing system, a decision structure generated from data about previous modification actions of multiple types that are performed on configuration settings affecting use of hardware components on computing devices and that are associated with subsequent changes in device operations of the computing devices, wherein the computing devices and the modification actions have a plurality of device and action attributes, and wherein the decision structure includes information about measured effects for a subset of the device and action attributes that are selected based on combinations of those device and action attributes being associated with measured effects of an indicated type in the subsequent changes; and

using, by the configured computing system, the decision structure to control ongoing operations for an additional computing device, including;

determining, by the configured computing system, and for each of multiple modification actions to possibly perform, a combination of device and action attributes in the decision structure that matches attributes of the additional computing device and the modification action, and using information about the measured effects for the determined combination to predict an effect of performing the modification action on the additional computing device;

selecting, by the configured computing system, one of the multiple modification actions based at least in part on the predicted effect of performing the selected one modification action on the additional computing device; and

initiating, by the configured computing system, performance of the selected one modification action on the additional computing device to modify one or more configuration settings that affect use of hardware components on the additional computing device.

27. The non-transitory computer-readable medium of claim 26 wherein the additional computing device is a mobile device, wherein the decision structure is a decision tree data structure, and wherein the stored contents include software instructions that, when executed, program the one or more computing systems to perform the automated operations.

28. A system comprising:

one or more hardware processors of one or more computing systems; and

one or more memories with stored instructions that, when executed, cause the system to perform automated operations including at least;

obtaining decision information generated from data about previous modification actions of multiple types that are performed on configuration settings affecting use of hardware components on computing devices and that are associated with subsequent changes in device operations of the computing devices, wherein the computing devices and the modification actions have a plurality of device and action attributes, and wherein the decision structure includes information about measured effects for a subset of the device and action attributes that are selected based on combinations of those device and action attributes being associated with measured effects of an indicated type in the subsequent changes; and

using the obtained decision information to control ongoing operations for an additional computing device, including;

determining, for each of multiple modification actions to possibly perform, a combination of device and action attributes in the obtained decision information that matches attributes of the additional computing device and the modification action, and using information about the measured effects for the determined combination to predict an effect of performing the modification action on the additional computing device;

selecting one of the multiple modification actions based at least in part on the predicted effect of performing the selected one modification action on the additional computing device; and

initiating performance of the selected one modification action on the additional computing device to modify one or more configuration settings that affect use of hardware components on the additional computing device.

29. The system of claim 28 wherein the additional computing device is a mobile device, wherein the obtained decision information is encoded in a decision tree data structure, and wherein the stored instructions include software instructions that, when executed, program the one or more computing systems to perform the automated operations.