Method and apparatus for visualizing and interactively manipulating profile data

• US 9,576,382 B2
• Filed: 08/05/2013
• Issued: 02/21/2017
• Est. Priority Date: 09/30/2008
• Status: Active Grant

• Current Assignee: Nintendo Company Limited
• Original Assignee: Nintendo Company Limited
• Inventors: Rabin, Steve
• Primary Examiner(s): Wang, Jin-Cheng
• Application Number:
  US13/959,476
• Publication Number:
  US 20130326416A1
• Time in Patent Office: 1,296 Days
• Field of Search: 345/440, 345/440.1, 345/440.2, 345/441, 345/442, 345/443, 717/125, 715/835
• US Class Current: 1/1
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3664 Environments for testing or debugging software
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06F 3/04817 using icons graphical or visual programming using iconic symbols G06F8/34
• CPC Class Codes: G06T 11/203 Drawing of straight lines or curves
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: G06T 11/40 Filling a planar surface by adding surface attributes, e.g. colour or texture
• CPC Class Codes: G09G 5/20 Function-generator circuits, e.g. circle generators line or curve smoothing circuits
• CPC Class Codes: H02K 15/063 Windings for large electric machines, e.g. bar windings windings consisting of cables H02K15/065
• CPC Class Codes: H02K 3/48 in slots

Abstract

Profile data collected through sampling is transformed and visually and interactively manipulated and displayed. In one scenario, the profile data is collected through statistical profiling of a program for a video game. The game program profile data is visually and interactively manipulated and displayed to visually present a dynamic behavior of the video game that shows correlation of impacts that the functions of the program have on the video game'"'"'s performance. This allows the developer to identify inefficient sections of the program for optimization.

Claims

1. A method transforming data of performance measurements from execution of a program having plural functions and manipulating the transformed data to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the method comprising:

executing the plural functions on at least one processing device;

collecting the data of performance measurements from the plural functions to create profile data;

automatically generating, using at least one processor, iconic visual representations, depicting a preview of function performance over a duration of the profile data in the form of a mini-graph, based on the collected performance measurement data, each iconic visual representation corresponding to one of the plural functions and the function performance indicative of, at least, a percentage of execution time spent in a respective function;

using the at least one processor, causing the iconic visual representations to be displayed; and

in response to user selection of a function corresponding to one of the iconic visual representations, using the at least one processor to graphically display actual performance of the corresponding executing function.

2. The method of claim 1, wherein each iconic visual representation is generated using non-linear scaling.

3. The method of claim 2, wherein each iconic visual representation is generated using exponential scaling.

4. The method of claim 1, wherein the function performance is indicative of a percentage of execution time of a respective function over a frame of execution.

5. The method of claim 1, wherein each iconic visual representation is sorted based on the percentage of execution time spent in the respective function.

6. The method of claim 1, wherein the iconic visual representations are displayed in a first window of a display and the actual performance of the corresponding executing function is displayed in a second window of the display.

7. The method of claim 6, wherein selection of at least one iconic visual representation in the first window of the display causes the second window of the display to update to display the actual performance of the executing function corresponding to the selected iconic visual representation.

8. A method transforming data of performance measurements from execution of a program having plural functions and manipulating the transformed data to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the method comprising:

executing the plural functions on at least one processing device;

collecting the data of performance measurements from the plural functions to create profile data;

automatically generating, using at least one processor, function indicators based on the collected performance measurement data, each function indicator corresponding to one of the plural functions;

automatically generating, using the at least one processor, iconic visual representations, depicting a preview of function performance over a duration of the profile data in the form of a thumbnail, based on the collected performance measurement data, each iconic visual representation corresponding to one of the plural functions and the function performance indicative of, at least, a percentage of execution time spent in a respective function; and

using the at least one processor, causing the function indicators and the iconic visual representations to be displayed simultaneously such that each iconic visual representation is displayed to visually correlate with the function indicator that corresponds to the same function as that iconic visual representation.

9. The method of claim 8, further comprising:

in response to user selection of one of the function indicators, using the at least one processor to cause a performance graph of the function corresponding to the selected function indicator to be displayed based on the collected performance measurement data, the performance graph representing a graph of actual performance values of the corresponding function.

10. A method transforming data of performance measurements from execution of a program having plural functions and manipulating the transformed data to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the method comprising:

executing the plural functions on at least one processing device;

collecting the data of performance measurements from the plural functions;

automatically generating, using at least one processor, a performance graph of a selected function based on the collected performance measurement data, the selected function being one of the plural functions, and the performance graph representing a graph of actual performance values of the selected function, the actual performance values indicative of, at least, a percentage of execution time spent in a respective function;

automatically generating, using the at least one processor, a margin band of the selected function based on the collected performance measurement data, the margin band having a thickness representing a margin of error of the collected performance measurement data associated with the corresponding function and tracking the performance graph such that the margin band thickness smoothens to correspond to a change in the performance graph; and

using the at least one processor, causing the performance graph and the margin band to be simultaneously displayed.

11. The method of claim 10, wherein the margin band is generated to track a moving statistical representation of the performance graph over a time window when a change in the performance graph over the time window is within a predetermined percentage of the margin of error.

12. The method of claim 10, further comprising:

automatically generating, using the at least one processor, function indicators based on the collected performance measurement data, each function indicator corresponding to one of the plural functions, each function indicator being user selectable, and the selected function being a function corresponding to a function indicator selected by the user; and

using the at least one processor, causing the function indicators to be displayed.

13. The method of claim 12, wherein the margin band is displayed to visually correlate with the selected function indicator.

14. The method of claim 12, further comprising:

generating by the processor one or more iconic visual representations based on the collected performance measurement data, each iconic visual representation corresponding to one of the plural functions and providing an iconic graphical representation of execution performance of the function corresponding to that iconic visual representation; and

using the at least one processor, causing the iconic visual representations to be displayed such that each iconic visual representation visually correlates with the function indicator that corresponds to the same function as that iconic visual representation.

15. The method of claim 14, wherein each iconic graphical representation is generated using non-linear scaling.

16. The method of claim 10, wherein the margin band tracks a moving average of the performance graph within a predetermined window when a change in the performance graph within the predetermined window is within a predetermined percentage of the margin of error.

17. The method of claim 10, wherein the margin band tracks the performance graph substantially exactly when a change in the performance graph within a predetermined window is greater than a predetermined percentage of the margin of error.

18. The method of claim 17, wherein the margin band thickness decreases to a size substantially corresponding to a size of the performance graph when the change in the performance graph within the predetermined window is greater than the predetermined percentage of the margin of error.

19. The method of claim 10, wherein the margin band tracks the performance graph substantially exactly when there are immediate changes in the performance graph.

20. The method of claim 10, wherein the performance graph of the selected function can extend beyond portions outside of the margin band.

21. A profile analyzer structured to create profile data, to transform data of performance measurements from execution of a program having plural functions, and to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the profile analyzer comprising:

a processor structured to generate iconic visual representations, depicting a preview of function performance over a duration of the profile data in the form of a mini-graph, based on the performance measurement data, each iconic visual representation corresponding to one of the plural functions and the function performance indicative of, at least, a percentage of execution time spent in a respective function; and

a display controller structured to control a display device

to display the iconic visual representations, and

in response to user selection of a function corresponding to one of the iconic visual representations, to graphically display actual performance of the corresponding executing function.

22. The profile analyzer of claim 21, wherein the processor is structured to sort the iconic visual representations based on the performances of the corresponding plural functions.

23. A profile analyzer structured to create profile data, to transform data of performance measurements from execution of a program having plural functions, and to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the profile analyzer comprising:

a processor structured to

generate function indicators based on the performance measurement data, each function indicator corresponding to one of the plural functions, and

generate iconic visual representations, depicting a preview of function performance over a duration of the profile data in the form of a thumbnail, based on the performance data, each iconic visual representation corresponding to one of the plural functions and the function performance indicative of, at least, a percentage of execution time spent in a respective function; and

a display controller structured to control a display device to simultaneously display the function indicators and the iconic visual representations such that each iconic visual representation is displayed to visually correlate with the function indicator that corresponds to the same function as that iconic visual representation.

24. The profile analyzer of claim 23, wherein the processor is structured to generate each iconic visual representation using non-linear scaling.

25. The profile analyzer of claim 23, wherein the display controller is structured to control the display device to display, in response to a user selection of one of the function indicators, a performance graph of the function corresponding to the selected function indicator based on the performance measurement data, the performance graph representing a graph of actual performance values of the corresponding function.

26. The profile analyzer of claim 25, wherein the display controller is structured to control the display device to display a margin band of the corresponding function to be displayed with the performance graph based on the performance measurement data, the margin band tracking the performance graph, and a thickness of the margin band representing a margin of error of the performance measurement data associated with the corresponding function.

27. The profile analyzer of claim 26, wherein the display controller is structured to control the display device to visually correlate the margin band with the selected function indicator.

28. A profile analyzer structured to transform data of performance measurements from execution of a program having plural functions and to visually present a dynamic behavior of the program that shows correlation of impacts that the plural functions of the program have on execution performance of the executing program, the profile analyzer comprising:

a processor structured to

generate a performance graph of a selected function based on the performance measurement data, the selected function being one of the plural functions, and the performance graph representing a graph of actual performance values of the selected function, the actual performance values indicative of, at least, a percentage of execution time spent in a respective function, and

generate a margin band of the selected function based on the performance measurement data, the margin band having a thickness representing a margin of error of the performance measurement data associated with the corresponding function and tracking the performance graph such that the margin band thickness smoothens to correspond to a change in the performance graph; and

a display controller structured to control a display device to simultaneously display the performance graph and the margin band.

29. The profile analyzer of claim 28, wherein the processor is structured to generate the margin band to track a moving statistical representation of the performance graph over a time window when a change in the performance graph over the time window is within a predetermined percentage of the margin of error.

30. The profile analyzer of claim 28,

wherein the processor is structured to generate function indicators based on the performance measurement data, each function indicator corresponding to one of the plural functions, each function indicator being user selectable, and the selected function being a function corresponding to a function indicator selected by the user, and

wherein the display controller is structured to control the display device to display the function indicators.

31. The profile analyzer of claim 30, wherein the display controller is structured to control the display device to visually correlate the margin band with the selected function indicator.

32. The profile analyzer of claim 30,

wherein the processor is structured to generate one or more iconic visual representations based on the performance measurement data, each iconic visual representation corresponding to one of the plural functions and providing an iconic graphical representation of execution performance of the function corresponding to that iconic visual representation, and

wherein the display controller is structured to control the display device to display the iconic visual representations such that each iconic visual representation visually correlates with the function indicator that corresponds to the same function as that iconic visual representation.

33. The profile analyzer of claim 32, wherein the processor is structured to generate each iconic graphical representation using non-linear scaling.