Apparatus, method and product for optimizing software system workload performance scenarios using multiple criteria decision making

• US 7,716,151 B2
• Filed: 02/09/2007
• Issued: 05/11/2010
• Est. Priority Date: 02/13/2006
• Status: Active Grant

• Current Assignee: Infosys Limited
• Original Assignee: Infosys Technologies Limited (Infosys Limited)
• Inventors: Ganesan, Rajeshwari, Tiwari, Nidhi
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): KIM, DAVID H
• Application Number:
  US11/704,293
• Publication Number:
  US 20070192266A1
• Time in Patent Office: 1,187 Days
• Field of Search: 706/45, 706/18, 703/21, 703/22
• US Class Current: 706/45
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

The present technique is an apparatus and method evaluating software performance. The method identifies performance scenarios using a knowledge base and selects the performance scenarios from a context module using operational laws. The system analyzes performance criticality of an application workload. Furthermore, the system comprises a context module that ranks the performance criterions depending on the input and the knowledge base stores and retrieves the performance criterions using the operational laws. The knowledge base comprises the criterions and their relevant ranks based of the application context. The present technique automates classification of performance criterions into benefit and cost categories with the usage of the operational laws.

Claims

1. A method of evaluating performance of at least one software system, the method comprising:

using a computing device, identifying a plurality of performance criterion using a plurality of operational laws, wherein a plurality of operational laws comprise a Little Law equation and a utilization law, wherein a utilization law is device throughput multiplied by service demand, whereby service demand for the devices remain constant;

gathering from the user a set of application details, which are associated with an application scenario;

mapping the application details with a patterning module that uses a knowledge base to rank the criteria to reflect the requirements of the application scenarios based on context details of user input; and

using a computing device, selecting a plurality of performance scenarios for the at least one software system based on at least one of the plurality of performance criterion using at least one Multiple Criteria Decision Making (MCDM) technique;

wherein the MCDM technique comprises computing a plurality of ranks of the application scenarios for each of a cost criteria and a benefit criteria individually, aggregating the application scenario ranks for the benefit criteria and cost criteria, and computing an overall criticality rank for the application scenarios by taking the ratio of the benefit criteria and cost criteria ranks.

2. The method as recited in claim 1, further comprising capturing the plurality of performance criterion to select the plurality of performance scenarios for at least one input using the plurality of operational laws.

3. The method as recited in claim 1, further comprising mapping the plurality of performance criterion with the at least one input using a knowledge base.

4. The method as recited in claim 3, wherein creating and representing the knowledge base indicative of the plurality of performance criterion using the plurality of operational laws.

5. The method as recited in claim 1, further comprising first ranking a plurality of scenarios of each of a plurality of quantitative or qualitative performance criterion of the at least one software system to identify the plurality of performance scenarios using a ranking module.

6. The method as recited in claim 1, further comprising ranking to combine a plurality of performance scenario ranks for each of the plurality of performance criterion selected as at least one benefit or at least one cost criteria using at least one Multiple Criteria Decision Making (MCDM) technique.

7. The method as recited in claim 6, wherein ranking the plurality of performance scenario for the at least one input comprises using a cost benefit analysis.

8. The method as recited in claim 7, wherein the at least one MCDM technique is used to evaluate and analyze performance of the at least one software system of at least one scenario.

9. The method as recited in claim 6, wherein the at least one qualitative criteria comprises a plurality of scale of intensities selected as low, or medium or high value.

10. The method as recited in claim 9, wherein comparing the low or medium or high value from a multiple set of the plurality of scale of intensities for computing at least one rank.

11. The method as recited in claim 10, wherein the at least one rank of each of the plurality of scale of intensities is displayed using at least one historical data of the knowledge base.

12. The method as recited in claim 1, further comprising listing the plurality of performance criterion and storing into the knowledge base using the plurality of operational laws.

13. The method as recited in claim 1, further comprising retrieving the plurality of performance criterion depending on the at least one input of a plurality of users.

14. The method as recited in claim 1, further comprising classifying the plurality of performance criterion into at least one benefit criteria or at least one cost criteria or both using the plurality of operational laws.

15. The method as recited in claim 14, wherein the at least one benefit criteria comprise a plurality of high value criteria and the at least one cost criteria comprise a plurality of low value criteria.

16. The method as recited in claim 14, further comprising deriving at least one benefit rank of the at least one benefit criteria and at least one cost rank of the at least one cost criteria using the at least one MCDM technique.

17. The method as recited in claim 16, further comprising comparing the at least one benefit criteria and the at least one cost criteria with multiple set of the at least one cost criteria and the at least one benefit criteria using one of the at least one MCDM technique.

18. A system for evaluating performance of at least one software system, the system comprising:

a computer processor adapted to allow modules to function in a system;

an input module adapted to obtain a plurality of performance criterion depending on at least one input and storing into a knowledge base using a plurality of operational laws, wherein a plurality of operational laws comprise a Little Law equation and a utilization law, wherein a utilization law is device throughput multiplied by service demand, whereby service demand for the devices remain constant;

a gathering module adapted to gather from the user a set of application details, which are associated with an application scenario;

a patterning module that uses the knowledge base to rank the criteria to reflect the requirements of the application scenarios based on context details of user input;

a selection module adapted to select a plurality of performance scenarios for the at least one software system depending on at least one of the plurality of performance criterion using at least one Multiple Criteria Decision Making (MCDM) technique; and

an output module adapted to display the plurality of performance scenarios based on the at least one input and the plurality of performance criterion and at least one rank using the at least one MCDM technique,

wherein the MCDM technique comprises computing a plurality of ranks of the application scenarios for each of a cost criteria and a benefit criteria individually, aggregating the application scenario ranks for the benefit criteria and cost criteria, and computing an overall criticality rank for the application scenarios by taking the ratio of the benefit criteria and cost criteria ranks.

19. The system as recited in claim 18, further comprising mapping the plurality of performance criterion to the at least one input and listing the plurality of performance criterion with at least one rank.

20. The system as recited in claim 19, wherein the knowledge base adapted to store and retrieve the plurality of performance criterion with the at least one rank.

21. The system as recited in claim 18, further comprising at least one historical data of the knowledge base adapted to display the at least one rank as required by at least one user.

22. The system as recited in claim 18, further comprising a plurality of operational laws adapted to classify the plurality of performance criterion into at least one cost criteria and at least one benefit criteria.

23. The system as recited in claim 18, further comprising the at least one MCDM technique adapted to relatively first rank each of the plurality of performance criterion.

24. The system as recited in claim 18, further comprising the at least one MCDM technique adapted to relatively rank each of a plurality of scenarios, wherein the at least one MCDM technique calculates a ratio by dividing at least one benefit rank with at least one cost rank.

25. A computer program product tangibly embodying a plurality of instructions, which when executed by a computer processor cause said processor to carry out steps for evaluating performance of at least one software system, the computer program product comprising:

program code adapted to select a plurality of performance criterion using a plurality of operational laws, wherein a plurality of operational laws comprise a Little Law equation and a utilization law, wherein a utilization law is device throughput multiplied by service demand, whereby service demand for the devices remain constant;

program code adapted to gather from the user a set of application details, which are associated with an application scenario program code adapted to create and represent a knowledge base indicative of the plurality of performance criterion using the plurality of operational laws;

program code adapted to map the plurality of performance criterion to the knowledge base to reflect the requirements of the application scenario based on context details of user input; and

program code adapted to rank a plurality of scenarios of the at least one software system depending on the mapping of the plurality of performance criterion using the at least one Multiple Criteria Decision Making (MCDM) technique

wherein the MCDM technique comprises computing a plurality of ranks of the application scenarios for each of a cost criteria and a benefit criteria individually, aggregating the application scenario ranks for the benefit criteria and cost criteria, and computing an overall criticality rank for the application scenarios by taking the ratio of the benefit criteria and cost criteria ranks.

26. The computer program product as recited in claim 25, further comprising program code adapted to classify the plurality of performance criterion as at least one benefit criteria and at least one cost criteria using the plurality of operational laws.

27. The computer program product as recited in claim 25, further comprising program code adapted to providing at least one rank of the plurality of performance criterion using the at least one MCDM technique.

28. The computer program product as recited in claim 26, further comprising program code adapted to display at least one performance critical rank of the plurality of scenarios in descending order.