Optimization of a message handling system

• US 8,195,790 B2
• Filed: 03/09/2006
• Issued: 06/05/2012
• Est. Priority Date: 03/24/2005
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Braithwaite, Kevin Anthony, Dunn, Tim, Garza, Jose Emir
• Primary Examiner(s): Fan, Hua
• Application Number:
  US11/371,746
• Publication Number:
  US 20060230177A1
• Time in Patent Office: 2,280 Days
• Field of Search: 709223-224
• US Class Current: 709/224
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways

Abstract

A data processing method for optimizing one or more parameters of a message handling system comprises defining a test workload comprising a plurality of messages, simulating the normal operation of the message handling system, processing at least a portion of the test workload, changing one or more parameters of the message handling system, monitoring the performance of the message handling system, and generating an optimization of at least one parameter based upon the performance of the message handling system.

Claims

1. A data processing method for optimizing one or more parameters of a message handling system, said method comprising:

defining a test workload comprising a plurality of messages, wherein the test workload is defined exclusively by the message handling system by monitoring normal operation of the message handling system;

simulating the normal operation of the message handling system;

processing at least a portion of the test workload;

changing one or more parameters of the message handling system;

monitoring performance of the message handling system with respect to the processing of at least a portion of the test workload;

generating an optimization of at least one parameter based upon the performance of the message handling system with respect to the processing of at least a portion of the test workload, wherein the simulating is performed by a processing component of the message handling system, and wherein the normal operation of the message handling system is stopped upon the message handling system entering a test mode that includes processing of the at least a portion of the test workload;

sending the plurality of messages of the test workload to an input queue;

monitoring a depth of the input queue, wherein a rate of placing messages on the input queue is increased if there is potential for more messages to be placed on the input queue; and

creating a draining process to drain the plurality of messages from an output queue, wherein changing one or more parameters of the message handling system and monitoring the performance of the message handling system include using a genetic algorithm to modify the one or more parameters of the message handling system in random fashion and test a result with a learning process.

2. The method according to claim 1, further comprising receiving a signal to trigger generating the optimization.

3. The method according to claim 2, wherein the signal is generated automatically by the message handling system.

4. The method according to claim 1, wherein the test workload includes a defined time period over which the at least a portion of the test workload is processed.

5. The method according to claim 1, wherein each of the plurality of messages in the test workload is designated by a type.

6. The method according to claim 5, wherein simulating the normal operation of the message handling system and processing at least a portion of the test workload are executed for each message type in turn.

7. The method according to claim 1, wherein simulating the normal operation of the message handling system and processing at least a portion of the test workload are executed for the entire test workload.

8. The method according to claim 1, wherein changing one or more parameters of the message handling system and monitoring the performance of the message handling system comprise changing only a single parameter and monitoring effects of the change of the single parameter.

9. The method of claim 1, wherein the test workload is representative of a normal workload of the message handling system and the performance corresponds to the processing of the at least a portion of the test workload.

10. A message handling system, comprising a non-transitory computer readable storage medium and a plurality of interconnected processing components for maintaining, communicating and processing messages, the plurality of interconnected processing components including a facility exclusively defining a test workload comprising a plurality of messages, wherein the test workload is defined exclusively by the message handling system by monitoring normal operation of the message handling system, storing the test workload for subsequent update to the test workload, simulating the normal operation of the message handling system, processing at least a portion of the test workload, changing at least one parameter of the message handling system, monitoring performance of the message handling system with respect to the processing of at least a portion of the test workload, generating an optimization of at least one parameter based upon the performance of the message handling system with respect to the processing of at least a portion of the test workload, wherein the simulating is performed by a processing component of the message handling system, and wherein the normal operation of the message handling system is stopped upon the message handling system entering a test mode that includes processing of the at least a portion of the test workload;

sending the plurality of messages of the test workload to an input queue;

monitoring a depth of the input queue, wherein a rate of placing messages on the input queue is increased if there is potential for more messages to be placed on the input queue; and

creating a draining process to drain the plurality of messages from an output queue, wherein changing at least one parameter of the message handling system and monitoring the performance of the message handling system include using a genetic algorithm to modify the at least one parameter of the message handling system in random fashion and test a result with a learning process.

11. The system according to claim 10, wherein the facility is arranged to define the test workload in response to receipt of a signal.

12. The system according to claim 11, wherein the signal is generated automatically by the message handling system.

13. The system according to claim 10, wherein the test workload includes a defined time period over which the at least a portion of the test workload is processed.

14. The system according to claim 10, wherein each of the plurality of messages in the test workload is designated by a type.

15. The system according to claim 14, wherein the facility, when simulating the normal operation of the message handling system, and processing at least a portion of the test workload, is arranged to repeat these operations for each message type in turn.

16. A computer program product on a non-transitory computer readable storage medium comprising instructions for optimizing one or more parameters of a message handling system, said instructions comprising defining a test workload comprising a plurality of messages, wherein the test workload is defined exclusively by the message handling system based on monitoring normal operation of the message handling system, storing the test workload, simulating the normal operation of the message handling system, processing at least a portion of the test workload, changing one or more parameters of the message handling system, monitoring performance of the message handling system with respect to the processing of at least a portion of the test workload, generating an optimization of at least one parameter based upon the performance of the message handling system with respect to the processing of at least a portion of the test workload, wherein the simulating is performed by a processing component of the message handling system, and wherein the normal operation of the message handling system is stopped upon the message handling system entering a test mode that includes processing of the at least a portion of the test workload;

sending the plurality of messages of the test workload to an input queue;

monitoring a depth of the input queue, wherein a rate of placing messages on the input queue is increased if there is potential for more messages to be placed on the input queue; and

creating a draining process to drain the plurality of messages from an output queue, wherein changing one or more parameters of the message handling system and monitoring the performance of the message handling system include using a genetic algorithm to modify the one or more parameters of the message handling system in random fashion and test a result with a learning process.

17. The computer program product of claim 16, wherein each of the plurality of messages in the test workload is designated by a type, and wherein the simulating the normal operation of the message handling system and the processing at least a portion of the test workload are performed for each message type in turn.

18. The computer program product of claim 16, wherein the changing one or more parameters of the message handling system and the monitoring the performance of the message handling system are performed recursively.