Intelligent work load manager

Abstract

An apparatus comprising a processor configured to monitor at a recurring interval a plurality of message queues and a plurality of server sessions that access the message queues, determine, at each occurrence of the interval, an appropriate number of server sessions per queue that achieves acceptable load balance, and redistribute the server sessions among the message queues to match the appropriate number of server session per queue upon determining the appropriate number of server sessions and prior to a subsequent occurrence of the interval.

What is claimed is: 
     
       1. An apparatus comprising a processor configured to:
 monitor at a recurring interval a plurality of message queues and a plurality of server sessions that access the message queues;   determine, at each occurrence of the interval, an appropriate number of server sessions per queue that achieves acceptable load balance; and   redistribute the server sessions among the message queues to match the appropriate number of server session per queue upon determining the appropriate number of server sessions and prior to a subsequent occurrence of the interval.   
     
     
       2. The apparatus of  claim 1 , wherein the processor is configured to maintain the server sessions for the message queues if the message queues each have a depth beyond a threshold depth value for determining whether a message queue is relatively lightly loaded. 
     
     
       3. The apparatus of  claim 1 , wherein the processor is configured to leave a minimum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       4. The apparatus of  claim 1 , wherein the processor is configured to move a maximum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       5. The apparatus of  claim 1 , wherein the servers' sessions are activation specification sessions implemented using an International Business Machines (IBM) Websphere application program. 
     
     
       6. A method for intelligently managing a work load comprising:
 monitoring at a recurring interval a plurality of message queues and a plurality of server sessions that access the message queues;   determining, at each occurrence of the interval, an appropriate number of server sessions per queue that achieves acceptable load balance; and   redistributing the server sessions among the message queues to match the appropriate number of server session per queue upon determining the appropriate number of server sessions and prior to a subsequent occurrence of the interval.   
     
     
       7. The method of  claim 6 , further comprising maintaining the server sessions for the message queues if the message queues each have a depth beyond a threshold depth value for determining whether a message queue is relatively lightly loaded. 
     
     
       8. The method of  claim 6 , further comprising leaving a minimum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       9. The method of  claim 6 , further comprising moving a maximum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       10. The method of  claim 6 , wherein the servers' sessions are activation specification sessions implemented using an International Business Machines (IBM) Websphere application program. 
     
     
       11. A non-transitory computer-readable medium having computer usable instructions stored thereon for execution by a processor, wherein the instructions cause the processor to:
 monitor at a recurring interval a plurality of message queues and a plurality of server sessions that access the message queues;   determine, at each occurrence of the interval, an appropriate number of server sessions per queue that achieves acceptable load balance; and   redistribute the server sessions among the message queues to match the appropriate number of server session per queue upon determining the appropriate number of server sessions and prior to a subsequent occurrence of the interval.   
     
     
       12. The non-transitory computer-readable medium of  claim 11 , wherein the instructions further cause the processor to maintain the server sessions for the message queues if the message queues each have a depth beyond a threshold depth value for determining whether a message queue is relatively lightly loaded. 
     
     
       13. The non-transitory computer-readable medium of  claim 11 , wherein the instructions further cause the processor to leave a minimum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       14. The non-transitory computer-readable medium of  claim 11 , wherein the instructions further cause the processor to move a maximum number of sessions per queue upon redistributing the server sessions among the message queues. 
     
     
       15. The non-transitory computer-readable medium of  claim 11 , wherein the servers' sessions are activation specification sessions implemented using an International Business Machines (IBM) Websphere application program. 
     
     
       16. A method for intelligently managing a work load, the method comprising:
 monitoring a plurality of message queues and a plurality of server resources available to service a plurality of server sessions that access the message queues;   determining an appropriate number of server resources that achieves acceptable load balance; and   redistributing a current distribution of the server resources among the message queues to better match an appropriate number of server sessions per queue, based on determining the appropriate number of server resources that achieves acceptable load balance.   
     
     
       17. A method according to  claim 16 , further comprising:
 maintaining the server sessions for the message queues if the message queues each have a depth beyond a threshold depth value for determining whether a message queue is relatively lightly loaded.   
     
     
       18. A method according to  claim 16 , further comprising:
 leaving a minimum number of sessions per queue upon redistributing the server resources.   
     
     
       19. A method according to  claim 16 , further comprising:
 moving a maximum number of sessions per queue upon redistributing the server resources among the message queues.   
     
     
       20. A non-transitory computer-readable medium having computer usable instructions stored thereon for execution by a processor, wherein the instructions cause the processor to:
 monitor a plurality of message queues and a plurality of server resources available to service a plurality of server sessions that access the message queues;   determine an appropriate number of server resources that achieves acceptable load balance; and   redistribute a current distribution of the server resources among the message queues to better match an appropriate number of server sessions per queue, based on determining the appropriate number of server resources that achieves acceptable load balance.   
     
     
       21. A non-transitory computer-readable medium according to  claim 20 , wherein the instructions further cause the processor to:
 maintain the server sessions for the message queues if the message queues each have a depth beyond a threshold depth value for determining whether a message queue is relatively lightly loaded.   
     
     
       22. A non-transitory computer-readable medium according to  claim 20 , wherein the instructions further cause the processor to:
 leave a minimum number of sessions per queue upon redistributing the server resources.   
     
     
       23. A non-transitory computer-readable medium according to  claim 20 , wherein the instructions further cause the processor to:
 move a maximum number of sessions per queue upon redistributing the server resources among the message queues.