Visual program runtime performance analysis

• US 6,189,142 B1
• Filed: 09/16/1998
• Issued: 02/13/2001
• Est. Priority Date: 09/16/1998
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Johnston, Gary Martin, Keren, Mila, Shaham-Gafni, Yael, Lindsey, Anthony Hayden
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): DAM, TUAN QUANG
• Application Number:
  US09/154,126
• Time in Patent Office: 881 Days
• Field of Search: 717/4, 717/1, 717/2, 714/35, 714/38, 714/39, 714/47, 714/45, 714/46, 345/333, 345/339, 345/326, 345/967, 345/968, 345/334, 345/348, 345/349, 702/182, 702/183, 702/186
• US Class Current: 717/125
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3428 Benchmarking
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06F 2201/88 Monitoring involving counting
• CPC Class Codes: G06F 8/34 Graphical or visual programming
• CPC Class Codes: Y10S 715/967 Visual or iconic programming

Abstract

A method, system, and computer program for providing runtime performance analysis in a visual programming environment. A performance data collection technique for visual programming is defined, which accounts for the specialized nature of visual programming, and the analysis is based on this collected data. The performance data is expressed in terms of elements familiar to the visual programmer, as opposed to using a source code approach. In a preferred embodiment, these elements comprise events, actions, connections, and attributes. The code implementing a visual program is instrumented according to this performance data collection technique, and performance information is gathered as the instrumented program executes. Preferably, the information gathered comprises execution counts and elapsed time per element.

Claims

1. In a computing system having a capability for component-based visual programming, computer readable code readable by said system for providing runtime performance analysis, comprising:

a subprocess for providing one or more component-based, visually-constructed programs;

a subprocess for instrumenting a selected one of said visually-constructed programs in order to collect runtime performance data, further comprising;

a subprocess for locating one or more elements of interest in said selected visually-constructed program, wherein said elements comprise instances of features of said selected visually-constructed program, said features comprising one more of;

(1) events to be signaled by components, (2) actions to be performed by components, (3) attributes of components, and (4) connections to be fired; and

a subprocess for adding a plurality of code hooks to said selected visually-constructed program for each of said located elements; and

a subprocess for gathering performance information for said selected visually-constructed program during runtime, using said instrumented program, further comprising;

a subprocess for executing one or more benchmarks against said instrumented program; and

a subprocess for logging an entry upon detection of any of said code hooks during operation of said subprocess for executing.

2. Computer readable code for providing runtime performance analysis according to claim 1, further comprising a subprocess for enabling a user of said computing system to select said elements of interest.

3. Computer readable code for providing runtime performance analysis according to claim 1, wherein particular ones of said logged entries comprise an indication that said located element corresponding to said detected code hook has been executed.

4. Computer readable code for providing runtime performance analysis according to claim 1, wherein particular ones of said logged entries comprise an indication of an elapsed execution time for said located element corresponding to said detected code hook.

5. Computer readable code for providing runtime performance analysis according to claim 1, wherein said computing system resides in a computer network.

6. Computer readable code for providing runtime performance analysis according to claim 1, wherein particular ones of said logged entries comprise an indication of an amount of data moved when one of said connections fired, said one of said connections being an attribute-to-attribute connection.

7. A system for providing runtime performance analysis in a computing system having a capability for component-based visual programming, comprising:

one or more component-based, visually-constructed programs;

means for instrumenting a selected one of said visually-constructed programs in order to collect runtime performance data, further comprising;

means for locating one or more elements of interest in said selected visually-constructed program, wherein said elements comprise instances of features of said selected visually-constructed program, said features comprising one or more of;

(1) events to be signaled by components, (2) actions to be performed by components, (3) attributes of components, and (4) connections to be fired; and

means for adding a plurality of code hooks to said selected visually-constructed program for each of said located elements; and

means for gathering performance information for said selected visually-constructed program during runtime, using said instrumented program, further comprising;

means for executing one or more benchmarks against said instrumented program; and

means for logging an entry upon detection of any of said code hooks during operation of said means for executing.

8. The system for providing runtime performance analysis according to claim 7, further comprising means for enabling a user of said computing system to select said elements of interest.

9. The system for providing runtime performance analysis according to claim 7, wherein particular ones of said logged entries comprise an indication that said located element corresponding to said detected code hook has been executed.

10. The system for providing runtime performance analysis according to claim 7, wherein particular ones of said logged entries comprise an indication of an elapsed execution time for said located element corresponding to said detected code hook.

11. The system for providing runtime performance analysis according to claim 7, wherein said computing system resides in a computer network.

12. The system for providing runtime performance analysis according to claim 7, wherein particular ones of said logged entries comprise an indication of an amount of data moved when one of said connections fired, said one of said connections being an attribute-to-attribute connection.

13. A method for providing runtime performance analysis in a computing system having a capability for component-based visual programming, comprising the steps of:

instrumenting a selected one of one or more component-based, visually-constructed programs in order to collect runtime performance data, further comprising the steps of;

locating one or more elements of interest in said selected visually-constructed program, wherein said elements comprise instances of features of said selected visually-constructed program, said features comprising one or more of;

(1) events to be signaled by components, (2) actions to be performed by components, (3) attributes of components, and (4) connections to be fired; and

adding a plurality of code hooks to said selected visually-constructed program for each of said located elements; and

gathering performance information for said selected visually-constructed program during runtime, using said instrumented program, further comprising the steps of;

executing one or more benchmarks against said instrumented program; and

logging an entry upon detection of any of said code hooks during operation of said executing step.

14. The method for providing runtime performance analysis according to claim 13, further comprising the step of enabling a user of said computing system to select said elements of interest.

15. The method for providing runtime performance analysis according to claim 13, wherein particular ones of said logged entries comprise an indication that said located element corresponding to said detected code hook has been executed.

16. The method for providing runtime performance analysis according to claim 13, wherein particular ones of said logged entries comprise an indication of an elapsed execution time for said located element corresponding to said detected code hook.

17. The method for providing runtime performance analysis according to claim 13, wherein said computing system resides in a computer network.

18. The method for providing runtime performance analysis according to claim 13, wherein particular ones of said logged entries comprise an indication of an amount of data moved when one of said connections fired, said one of said connections being an attribute-to-attribute connection.