Approach to monitor application states for self-managing systems

• US 7,487,494 B2
• Filed: 08/02/2004
• Issued: 02/03/2009
• Est. Priority Date: 08/02/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chieu, Trieu C, Kephart, Jeffrey Owen, Chan, Hoi Yeung
• Primary Examiner(s): Dam; Tuan Q
• Assistant Examiner(s): SMITH, CHENECA
• Application Number:
  US10/909,760
• Publication Number:
  US 20060026570A1
• Time in Patent Office: 1,646 Days
• Field of Search: None
• US Class Current: 717/127
• CPC Class Codes: G06F 2209/542 Intercept
• CPC Class Codes: G06F 8/316 Aspect-oriented programming techniques

Abstract

The present invention is directed to systems and methods that treat the monitoring of a software application as a separate and independent concern from the software application itself. The present invention utilizes aspect oriented programming to intercept, analyze, decompose the application states of a software application and then to link appropriate non-invasive constructs into the software application code to provide application state information to an autonomic manager through a plurality of sensors. The autonomic manager analyzes the application state information for compliance pre-defined parameters, and takes appropriate corrective action including modifying the software application states using a plurality of effectors. The sensors and effectors are linked as non-invasive constructs to the software application, and in aspect oriented programming constitute aspects. In order to prevent any adverse or undesirable effects to the software application as a result of the application state analysis and selected action, a safety mechanism is included in communication with the effectors.

Claims

1. A method for monitoring application states of software applications, the method comprising:

creating an autonomic manager comprising a plurality of sensors to receive data from a monitored software application and a plurality of effectors to affect actions within the monitored software application;

identifying linkage points within the monitored software application;

connecting sensors and effectors to the identified linkage points using non-invasive constructs disposed between the effectors and sensors and the monitored software application to form a cross-cutting layer comprising sensor point cuts and effector point cuts;

using the sensors to monitor application states associated with the monitored software application at run time for compliance with pre-defined performance parameters by diverting control to the sensor point cuts to collect application state information for analysis by the autonomic manager; and

using the effectors to execute corrective actions at run time when the application states are out of compliance by using the effector point cuts to initiate the corrective actions;

wherein the non-invasive constructs are external to the sensors, the effectors and the monitored software application and do not modify any source code associated with the monitored software application.

2. The method of claim 1, wherein the step of using the sensors to monitor the application states comprises:

monitoring values of application variables; and

monitoring application process sequences; and

the method further comprises analyzing the variables and sequences for compliance with the pre-defined performance parameters.

3. The method of claim 2, wherein the step of using the effectors to execute corrective actions further comprises modifying at least one of the values of the application variables and the process sequences.

4. The method of claim 1, wherein the sensor point cuts are disposed between the linkage points and the sensors and the effector point cuts are disposed between the linkage points and the effectors, and the method further comprises:

diverting control to an appropriate sensor point cut to collect application state data from the monitored application when a given linkage point in the monitored software application is reached;

analyzing the collected application state data in the autonomic manager; and

using the appropriate effector point cut to divert actions initiated by the autonomic manager to the given linkage point to execute the corrective actions.

5. The method of claim 1, wherein the step of creating the autonomic manager comprises using an architecture neutral programming language to create the autonomic manager.

6. The method of claim 5, wherein the architecture neutral programming language comprises JAVA.

7. The method of claim 1, further comprising providing the application state information to the autonomic manager using the non-invasive constructs.

8. The method of claim 1, further comprising inhibiting the corrective actions from adversely affecting operation of the software application.

9. The method of claim 8, wherein the step of inhibiting the corrective actions from adversely affecting operation comprises using a variable map to constrain the corrective actions within acceptable bounds as defined by the variable map.

10. The method of claim 9, further comprising using historical application state information to generate the variable map.

11. The method of claim 8, wherein the step of inhibiting the corrective actions from adversely affecting operation comprises using a heuristic function to constrain the corrective actions within acceptable bounds as defined by the heuristic function.

12. A computer readable medium containing a computer executable code that when read by a computer causes the computer to perform a method for monitoring application states of software applications, the method comprising:

creating an autonomic manager comprising a plurality of sensors to receive data from a monitored software application and a plurality of effectors to affect actions within the monitored software application;

identifying linkage points within the monitored software application;

connecting sensors and effectors to the identified linkage points using non-invasive constructs disposed between the effectors and sensors and the monitored software application to form a cross-cutting layer comprising sensor point cuts and effector point cuts;

using the sensors to monitor application states associated with the monitored software application at run time for compliance with pre-defined performance parameters by diverting control to the sensor point cuts to collect application state information for analysis by the autonomic manager; and

using the effectors to execute corrective actions at run time when the application states are out of compliance by using the effector point cuts to initiate the corrective actions;

wherein the non-invasive constructs are external to the sensors, the effectors and the monitored software application and do not modify any source code associated with the monitored software application.

13. The computer readable medium of claim 12, wherein the step of using the sensors to monitor the application states comprises:

monitoring values of application variables; and

monitoring application process sequences; and

the method further comprises analyzing the variables and sequences for compliance with the pre-defined performance parameters.

14. The computer readable medium of claim 13, wherein the step of using the effectors to execute corrective actions further comprises modifying at least one of the values of the application variables and the process sequences.

15. The computer readable medium of claim 12, wherein the sensor point cuts are disposed between the linkage points and the sensors and the effector point cuts are disposed between the linkage points and the effectors, and the method further comprises:

diverting control to an appropriate sensor point cut to collect application state data from the monitored application when a given linkage point in the monitored software application is reached;

analyzing the collected application state data in the autonomic manager; and

using the appropriate effector point cut to divert actions initiated by the autonomic manager to the given linkage point to execute the corrective actions.

16. The computer readable medium of claim 12, wherein the step of creating the autonomic manager comprises using an architecture neutral programming language to create the autonomic manager.

17. The computer readable medium of claim 16, wherein the architecture neutral programming language comprises JAVA.

18. The computer readable medium of claim 12, further comprising providing the application state information to the autonomic manager using the non-invasive constructs.

19. The computer readable medium of claim 12, further comprising inhibiting the corrective actions from adversely affecting operation of the software application.

20. The computer readable medium of claim 19, wherein the step of inhibiting the corrective actions from adversely affecting operation comprises using a variable map to constrain the corrective actions within acceptable bounds as defined by the variable map.

21. The computer readable medium of claim 20, further comprising using historical application state information to generate the variable map.

22. The computer readable medium of claim 19, wherein the step of preventing the corrective actions from adversely affecting operation comprises using a heuristic function to constrain the corrective actions within acceptable bounds as defined by the heuristic function.

23. A system for monitoring application states of software applications. the system comprising:

at least one software application comprising one or more application states and a plurality of linkage points;

an external entity comprising at least a processor, in communication with and independent of the software application, the external entity comprising a plurality of sensors to receive data from a monitored software application and a plurality of effectors to affect actions within the monitored software application; and

a plurality of non-invasive constructs connecting the sensors and effectors to the linkage points, the non-invasive constructs external to the sensors, the effectors and the software application and comprising a cross-cutting layer comprising sensor point cuts disposed between the linkage points and the sensors and effector point cuts disposed between the linkage points and the effectors.

24. The system of claim 23, wherein the external entity comprises an autonomic manager.

25. The system of claim 23, where in the external entity further comprises:

a processor to analyze the application state information obtained by the sensors for compliance with one or more pre-defined performance parameters and to select actions to be executed by the effectors based upon the analysis of the application state information.

26. The system of claim 23, further comprising a guarantee mechanism in communication with at least one of the non-invasive constructs to prevent actions selected by the external entity from adversely affecting operation of the software application.

27. The system of claim 23, wherein the system utilizes aspect oriented programming and each non-invasive construct comprises an aspect.

28. The system of claim 23, wherein the external entity monitors one or more of a plurality of user-selectable concerns associated with the software application.