Dynamically identifying performance anti-patterns

• US 10,346,283 B2
• Filed: 10/18/2017
• Issued: 07/09/2019
• Est. Priority Date: 12/22/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation, University College Dublin
• Original Assignee: International Business Machines Corporation
• Inventors: Altman, Erik R., Assem Aly Salama, Hitham Ahmed, Mitchell, Nicholas M., O'Sullivan, Patrick Joseph, Portillo Dominguez, Andres Omar, Sweeney, Peter F.
• Primary Examiner(s): Chavis, John Q
• Application Number:
  US15/787,046
• Publication Number:
  US 20180039560A1
• Time in Patent Office: 629 Days
• Field of Search: 717131
• CPC Class Codes: G06F 11/0706 the processing taking place on a specific hardware platform or in a specific software environment
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/3612 by runtime analysis performance monitoring G06F11/3466
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures

Abstract

Dynamically identifying performance anti-patterns in a software system is based on a set of documented symptoms that are evaluated in real-time. The evaluation is based on the observed system behavior and its comparison against the documented symptoms of different types of performance issues.

Claims

1. A computer-implemented method for improving the identification of performance anti-patterns in a software system, comprising:

using a processor device operably coupled with a global database for evaluating the software system for presence of an anti-pattern, wherein the anti-pattern is a behavioral trend suggestive of a particular performance issue that is likely to lead to system failure, by performing;

accessing the global database comprising a rules set for each functionality of interest to a software tester, wherein the rules set identifies an anti-pattern listed as a documented symptom for the particular performance issue, wherein the rules set comprises, for each anti-pattern;

a diagnostic tool to apply to detect the anti-pattern;

a rule to assess results of applying the diagnostic tool; and

a remedial action to perform when the anti-pattern is detected; and

iteratively performing the following at intervals;

selecting the rules set associated with the functionality of interest;

collecting data during a performance test run of the software system;

selecting the diagnostic tool associated with the anti-pattern from the rules set;

using the diagnostic tool, applying the rule on the collected data to detect presence of the anti-pattern;

based upon results of the application of the rule, revising the rule to improve identification of the anti-pattern in a subsequent performance test run; and

adding the collected data to the global database in association with the anti-pattern.

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

populating the global database with identified anti-patterns; and

associating the identified anti-patterns in the global database with the diagnostic tools used to identify the anti-patterns.

3. The computer-implemented method of claim 1 wherein detecting the presence of the anti-pattern comprises detecting the behavioral trend from a group consisting of:

response time degradation, a surge of blocked threads, an increase in garbage collection activity, and low resource utilization.

4. The computer-implemented method of claim 1 further comprising providing a time interval for collecting the data.

5. The computer-implemented method of claim 1 further comprising sharing the global database among multiple software systems.

6. The computer-implemented method of claim 1 wherein the computer-implemented method is run concurrently within an execution environment.

7. The computer-implemented method of claim 1 wherein selecting the diagnostic tool comprises selecting a thread and dump analyzer tool.

8. The computer-implemented method of claim 2 wherein associating the identified anti-patterns in the global database with the diagnostic tools comprises providing a pointer to the diagnostic tools.

9. The computer-implemented method of claim 1 further comprising expressing the anti-pattern in terms of business rules associated with a service level agreement.

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

identifying an additional anti-pattern; and

adding the additional anti-pattern to the global database.

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

selecting a system node for evaluating the software system; and

performing a remedial action as indicated in the global database, in association with the anti-pattern.

12. An information processing system for improving the identification of performance anti-patterns in a software system, comprising:

a global database listing a rules set for each functionality of interest to a software tester, wherein each rules set identifies an anti-pattern listed as a documented symptom for the particular performance issue, wherein the rules set comprises, for each anti-pattern;

a diagnostic tool to apply to detect the anti-pattern;

a rule to assess results of applying the diagnostic tool; and

a remedial action to perform when the anti-pattern is detected, wherein an anti-pattern is a behavioral trend suggestive of a particular performance issue that is likely to lead to system failure;

a processor device operably coupled with the global database; and

a memory operably coupled with the processor device, the memory comprising instructions that, when executed by the processor device, cause a computer for evaluating the software system for presence of an anti-pattern to;

iteratively perform the following at intervals;

selecting the rules set associated with the functionality of interest;

collecting data during a performance test run of the software system;

selecting the diagnostic tool associated with the anti-pattern from the rules set;

using the diagnostic tool applying the rule on the collected data to detect presence of the anti-pattern;

based upon results of the application of the rule, revising the rule to improve identification of the anti-pattern in a subsequent test run; and

adding the collected data to the global database in association with the anti-pattern.

13. The information processing system of claim 12 wherein the global database is shared among multiple systems.

14. The information processing system of claim 12 wherein the anti-patterns are expressed in terms of business rules associated with a service level agreement.

15. The information processing system of claim 12 further comprising:

selecting a system node for evaluating the software system, wherein the system node is one of;

an application and a function.

16. The information processing system of claim 12 wherein the diagnostic tool is a javacore file.

17. The information processing system of claim 12 wherein the behavioral trend is selected from a group consisting of:

response time degradation, a surge of blocked threads, an increase in garbage collection activity, and low resource utilization.

18. The information processing system of claim 12 wherein the global database includes a pointer to the diagnostic tool.

19. A computer program product for improving the identification of performance anti-patterns in a software system, the computer program product comprising:

a storage medium readable by a processing circuit and storing instructions for execution by the processing circuit coupled with a global database, for evaluating the software system for presence of an anti-pattern, wherein the anti-pattern is a behavioral trend suggestive of a particular performance issue that is likely to lead to system failure, the instructions comprising;

accessing the global database comprising a rules set for each functionality of interest to a software tester, wherein the rules set identifies an anti-pattern listed as a documented symptom for the particular performance issue, wherein the rules set comprises, for each anti-pattern;

a diagnostic tool to apply to detect the anti-pattern;

a rule to assess results of applying the diagnostic tool; and

a remedial action to perform when the anti-pattern is detected; and

iteratively performing the following at intervals;

selecting the rules set associated with the functionality of interest;

collecting data during a performance test run of the software system;

selecting the diagnostic tool associated with the anti-pattern from the rules set;

using the diagnostic tool applying the rule on the collected data to detect presence of the anti-pattern;

based upon results of the application of the rule, revising the rule to improve identification of the anti-pattern in a subsequent test run; and

adding the collected data to the global database in association with the anti-pattern.

20. The computer program product of claim 19 wherein the behavioral trend is selected from a group consisting of:

response time degradation, a surge of blocked threads, an increase in garbage collection activity, and low resource utilization.