[asterisk-bugs] [Asterisk 0011232]: Asterisk core not multitreading sip-to-sip calls

[noreply at bugs.digium.com]

The following issue has been RESOLVED. 
====================================================================== 
http://bugs.digium.com/view.php?id=11232 
====================================================================== 
Reported By:                coppens_b
Assigned To:                russell
====================================================================== 
Project:                    Asterisk
Issue ID:                   11232
Category:                   Core-General
Reproducibility:            always
Severity:                   feature
Priority:                   normal
Status:                     resolved
Asterisk Version:           1.4.13  
SVN Branch (only for SVN checkouts, not tarball releases): N/A  
SVN Revision (number only!):  
Disclaimer on File?:        N/A 
Request Review:              
Resolution:                 suspended
Fixed in Version:           
====================================================================== 
Date Submitted:             11-13-2007 10:06 CST
Last Modified:              11-14-2007 12:29 CST
====================================================================== 
Summary:                    Asterisk core not multitreading sip-to-sip calls
Description: 
Presently we are running asterisk in an sip-to-sip environment. In this
scenario, each call is using the following path:
G729 client -> [asterisk core] -> G729 client (no transcoding, with
canreinvite= yes).
For a certain amount of traffic, we have observed that CPU 0 is havely
loaded while the other CPU's 1,2 and 3 remain idle.
Setup:
DL360 G4  2* dual-core running Linux 2.6.18-8.el5
http://bugs.digium.com/view.php?id=1 SMP Fri Jan 26
14:15:21 EST 2007 i686 i686 i386 GNU/Linux
Asterisk 1.4.13 built by root @ ast2-gos2-be on a i686 running Linux on
2007-10-15 11:33:09 UTC

We retrieved the following stats for 120 simultenous sip-to-sip calls
05:42:26 PM  CPU   %user   %nice    %sys %iowait    %irq   %soft  %steal  
%idle    intr/s
05:42:31 PM  all    0.95    0.00    2.40    0.00    0.15    0.85    0.00  
95.65  11291.40
05:42:31 PM    0    3.60    0.00    9.60    0.00    0.40    3.60    0.00  
82.80  10280.60
05:42:31 PM    1    0.00    0.00    0.00    0.00    0.20    0.00    0.00 
100.00     10.80
05:42:31 PM    2    0.00    0.00    0.00    0.00    0.00    0.00    0.00 
100.00      0.00
05:42:31 PM    3    0.00    0.00    0.00    0.00    0.00    0.00    0.00 
100.00   1000.40

We are from the opinion that we should make the asterisk core
multithreaded.
This would allow us to handle *3 more calls.
====================================================================== 

---------------------------------------------------------------------- 
 russell - 11-14-07 12:29  
---------------------------------------------------------------------- 
You said that you have canreinvite=yes with G.729 on both sides.  In that
case, there isn't any RTP processing going on.  So, while there is a thread
for every single call, they sit idle most of the time.  The one thread
doing most of the work is the single incoming SIP signalling processing
thread.

In the CPU load output you posted, the first CPU is still 83% idle.  Have
you loaded it more than that?

If you would like to continue this discussion, let's do it on the
asterisk-dev mailing list.  We really just try to avoid using the bug
tracker as a discussion forum.  There are more people on the -dev list that
can contribute to this discussion. 

Issue History 
Date Modified   Username       Field                    Change               
====================================================================== 
11-14-07 12:29  russell        Note Added: 0073644                          
======================================================================