Joined: 30 Nov 2006 Posts: 614 Location: Washington DC
Ok.
So what we did was we deployed a webservice exposed as a webservice and deployed with additional instances 10(Because of heavy volume of data).performed a load test by pumping 500 requests at a given time and Worked fine without any timeouts no issues at all.
When we are monitoring the Memory on the LPAR it raised to nearly 200 MB and once all the request have been processed what we observed is the broker is not at all releasing the memory(We have WMB 6.0.0.3 ).
It is releasing the memory only when the broker is bounced.
What could be the problem.....how do we configure broker to release the unused memory.....
Ok.
So what we did was we deployed a webservice exposed as a webservice and deployed with additional instances 10(Because of heavy volume of data).performed a load test by pumping 500 requests at a given time and Worked fine without any timeouts no issues at all.
When we are monitoring the Memory on the LPAR it raised to nearly 200 MB and once all the request have been processed what we observed is the broker is not at all releasing the memory(We have WMB 6.0.0.3 ).
It is releasing the memory only when the broker is bounced.
What could be the problem.....how do we configure broker to release the unused memory.....
I believe it keeps the memory for a little while, because you could be receiving another msg on the queue that needs it. The memory should be released however over time. How long did you wait before coming to the realization that memory was not being released? _________________ MQ & Broker admin
Joined: 30 Nov 2006 Posts: 614 Location: Washington DC
had anyone successfully installed IE(Proxy servlet for the HTTP nodes) and found any drastic change in the performance of web services.
I have added like 30 additional instances as the volume of data is high they are getting timedout...may be Broker is the not the correct option of exposing it...
I would rather tell that broker works outstanding for asynchronous processing when it comes to synchronous and if it is less volume it handles in a good manner.
Joined: 30 Nov 2006 Posts: 614 Location: Washington DC
Anyone, please explain how the throughput varies by deploying the same messafe flows on one execution group with multiple threads vs deploying the same message flow on multiple execution groups.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum