« Back to VXML-API

RE: Max call optimization

Combination View Flat View Tree View
Threads [ Previous | Next ]
Max call optimization
Answer
7/18/12 10:53 AM
Good afternoon
 
I have 3745s and 3845s running TCL and VXML
We have noticed that nearby 200 concurrent calls 3745s destroys connections and 3845s do the same nearby 300 concurrent calls.
CPUs have never gone over 50%. Is there something I should do about memory or anything to be done to increase that number of concurrent calls?
 
Memory settings are default values.
 
Thanks!
 
Adolfo
Flag

RE: Max call optimization
Answer
9/6/12 10:14 AM as a reply to Adolfo Arizpe.
Hi Adolfo,

Sorry for missing out your query as we didn't get the mail alert for the same. With regard to your query about the performance issue for the 2 platforms, I will check our DE team and revert back to you.

Just wanted to keep you informed that 3745 platform has been EOLed.

Thanks,
Anusha
Flag

RE: Max call optimization
Answer
9/7/12 5:12 AM as a reply to Anusha Kannappan.
Hi Adolfo,

The maximum number of call varies from platform to platform. There is no specific API for controlling the same. As the higher number of concurrent calls requires more memory, we can just try playing less media so that less memory is being used. But we don't have any stats to tell what is the number of max call for what type of router.

Thanks,
Anusha
Flag

RE: Max call optimization
Answer
4/30/13 4:07 PM as a reply to Anusha Kannappan.
Thanks for your reply.
I wasn't until recently that this subject came up again. I am trying to increase the amount of concurrent calls, but at some point in 300 calls it stops playing some prompts.
I have used the following:
"ivr prompt memory"
"vxml memory tree" 
I have adjusted to see how can I perform better, without positive results
Also, I am running on default memory iomem, is there a setting that would help me do better?
There is very little information about these commands... 
 
Thanks again!
 
Adolfo
Flag

RE: Max call optimization
Answer
4/30/13 4:19 PM as a reply to Adolfo Arizpe.
Hi Adolfo,

Have you tried checking the memory and threads on the app server? Recently we were able to resolve a couple of scaling up issues by tweaking config on the Tomcat server that we were using.
 
Regards
Anupam
IBM Research
Flag