Analyzing performance of computing devices in usage scenarios

• US 8,850,172 B2
• Filed: 11/15/2010
• Issued: 09/30/2014
• Est. Priority Date: 11/15/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Dietrich, Aaron, Goyette, Sylvain, Lanning, Van Stephen
• Primary Examiner(s): DU, THUAN N
• Application Number:
  US12/946,229
• Publication Number:
  US 20120124363A1
• Time in Patent Office: 1,415 Days
• Field of Search: 713/1, 713/100, 718/104, 718/105
• US Class Current: 713/1
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• 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 2201/81 Threshold
• CPC Class Codes: G06F 2201/86 Event-based monitoring

Abstract

Techniques for conducting an automated analysis of operations carried out during the critical path for a usage scenario and suggesting ways in which the configuration of the computing device could be changed to affect performance of the computing device. Computing devices can be operated in a variety of usage scenarios and users may notice the performance of a computing device in certain usage scenarios more particularly. Critical path analysis of operations conducted in these usage scenarios can be used to identify a critical path of the usage scenario, from which changes that could be made to the computing device to affect performance could be identified. Once the changes that could be made are identified, suggestions can be made to the user, such that a user is able to make changes to the configuration to affect performance when the user has little knowledge about how to improve configurations.

Claims

1. A method comprising:

operating at least one programmed processor to carry out a set of acts, the at least one programmed processor programmed with processor-executable instructions that identify the set of acts, the set of acts comprising;

analyzing information about operations of a computing device generated during use of the computing device in a usage scenario;

identifying from the analyzed information at least one behavior pattern relating to performance of the computing device in the usage scenario; and

based at least in part on the at least one identified behavior pattern, presenting, on a display device, multiple suggestions that are ranked according to respective performance impacts associated with individual suggestions of the multiple suggestions for the computing device in the usage scenario, wherein the individual suggestions at least partially enable a user to effectuate at least one or more changes to a configuration of the computing device that would affect performance of the computing device in the usage scenario.

2. The method of claim 1, wherein:

the analyzing comprises comparing at least some of the information about operations to at least one known behavior pattern, where each at least one known behavior pattern is associated with a known configuration change that will affect performance of the computing device in the usage scenario, and

if a match is identified between the information about the operations and a first behavior pattern, at least one of the multiple ranked suggestions includes a first known configuration change associated with the first behavior pattern.

3. The method of claim 2, wherein:

comparing at least some of the information about operations to the at least one known behavior pattern comprises comparing at least some of the information about operations to first symptoms of a problem related to the first known behavior pattern, and

the first known configuration change comprises a solution associated with the first behavior pattern if at least some of the information about operations matches the first symptoms.

4. The method of claim 1, wherein:

the analyzing comprises analyzing chains of operations identified in the information about operations to identify at least one commonality between chains, the at least one commonality relating to execution of at least one software facility on the computing device; and

if at least one commonality is identified, the suggestion includes information regarding a change to the at least one software facility.

5. The method of claim 4, wherein identifying the at least one commonality comprises identifying the at least one commonality in a first trace and a second trace related to a first chain and a second chain identified in the information about operations.

6. The method of claim 4, wherein identifying the at least one commonality in the first chain and the second chain comprises identifying common thread identifiers, function names, and/or sequences of functions in a first chain and a second chain.

7. The method of claim 1, wherein:

the set of acts further comprises collecting tracing data upon occurrence of events during use of the computing device in the usage scenario, and

analyzing information about operations comprises analyzing the traces.

8. The method of claim 7, wherein the set of acts further comprises:

requesting collection of traces upon occurrence of the events based on an initialization file identifying how to monitor operation of the computing device in the usage scenario, the initialization file identifying the events for which traces are to be collected.

9. The method of claim 8, wherein:

the initialization file further identifies activity types and operations for the usage scenario, each of the activity types corresponding to at least one operation, each of the operations being related to an activity of a set of activities, and

the set of acts further comprises;

analyzing the traces to identify a set of operations that occurred during use of the computing device during the usage scenario;

analyzing the set of operations to determine the set of activities that form a part of the usage scenario on the computing device; and

for each activity of the set of activities, identifying behavior patterns that occurred during the activity.

10. The method of claim 1, wherein:

the analyzing comprises analyzing the information about operations to determine resources consumed by operations that were carried out by the computing device during the usage scenario and, for an operation that consumes more resources than an average, determining that a known solution exists for reducing resources consumed during the operation; and

at least one of the multiple ranked suggestions includes the known solution.

11. The method of claim 1, wherein at least one of the multiple ranked suggestions includes information regarding at least one first change to hardware of the computing device and/or at least one second change to software of the computing device.

12. At least one computer-readable storage medium encoded with computer-executable instructions that, when executed, cause a computer to carry out a method, the method comprising:

analyzing information about operations of a computing device generated during use of the computing device in a usage scenario to identify a set of operations executed during the usage scenario;

comparing information regarding the set of operations to at least one known behavior pattern, where each of the at least one known behavior pattern is associated with a known configuration change that will affect performance of the computing device in the usage scenario;

analyzing the set of operations to identify whether at least one commonality exists between operations of the set, the at least one commonality relating to execution of at least one software facility on the computing device; and

presenting, on a display device, multiple suggestions that are ranked according to performance impacts associated with individual ones of the multiple suggestions, wherein the individual suggestions at least partially enable a user to effectuate at least one or more changes to a configuration of the computing device that would affect performance of the computing device in the usage scenario, wherein at least one of the multiple ranked suggestions comprises;

if a match is identified between the set of operations and a first known behavior pattern, a first known configuration change associated with the first known behavior pattern; and

if at least one commonality exists between operations of the set, a change to the at least one software facility related to the at least one commonality.

13. The at least one computer-readable storage medium of claim 12, wherein:

comparing information regarding the set of operations to the at least one known behavior pattern comprises comparing information regarding operations of the set to symptoms of a problem for which there is a known solution, and

at least one of the multiple ranked suggestions includes the known solution if at least some of the information regarding the set of operations matches the symptoms.

14. The at least one computer-readable storage medium of claim 12, wherein identifying the at least one commonality in operations of the set comprises identifying common thread identifiers, function names, and/or sequences of functions in a first chain and a second chain.

15. The at least one computer-readable storage medium of claim 12, wherein:

the method further comprises generating traces upon occurrence of events during operation of the computing device in the usage scenario, and

analyzing information about operations comprises analyzing the traces.

16. The at least one computer-readable storage medium of claim 15, wherein:

the method further comprises requesting generation of traces upon occurrence of the events based on an initialization file identifying how to monitor operation of the computing device in the usage scenario, the initialization file identifying the events for which traces are to be collected, and

the initialization file further identifying activities of the usage scenario, each of the activities comprising at least one event, each of the events being related to an activity of a set of activities, and

the set of acts further comprises;

analyzing the traces to identify a set of events that occurred during use of the computing device during the usage scenario;

analyzing the set of events to determine the set of operations and, based on the set of operations, the set of activities that form a part of the usage scenario on the computing device; and

for each activity of the set of activities, identifying behavior patterns in operations of the activity.

17. A method for manufacturing computing devices comprising:

receiving a first specification of hardware and a second specification of software for a first computing device to be manufactured;

operating the first computing device in a usage scenario;

receiving a suggested change to be made to the first specification and/or second specification that would, if made to the first computing device, affect operation of the first computing device in the usage scenario, the suggested change being suggested based on an automated analysis of information about operations of the first computing device generated during the operating of the first computing device in the usage scenario;

making the suggested change to the first specification and/or the second specification; and

manufacturing multiple other computing devices according to the changed first specification and the changed second specification.

18. The method of claim 17, wherein receiving a suggested change comprises receiving an indication that operations of the first computing device, when operated in the usage scenario, include a behavior pattern indicative of a known issue affecting performance in the usage scenario and receiving an indication that a known configuration change exists that is associated with the behavior pattern;

and

wherein making the suggested change comprises making the known configuration change associated with the behavior pattern.

19. The method of claim 17, wherein receiving a suggested change comprises receiving an indication that operations of the first computing device, when operated in the usage scenario, include a behavior pattern related to a performance issue for which a solution is not known but that, if the performance issue is addressed, would improve performance of the first computing device;

and

wherein making the suggested change comprises making a change to the configuration to address the performance issue.

20. The method of claim 17, wherein the performance issue comprises an execution delay and/or a level of resource consumption, and

wherein the suggested change comprises a change that reduces the execution delay and/or the level of resource consumption.