[reSIProcate] StatelessProxy performance problem

[Robert Sparks]

Make sure whatever DNS names you are using actually resolve in DNS
(especially localhost).

RjS

On May 16, 2005, at 11:32 AM, Cullen Jennings wrote:

>
>  not sure what is going wrong here but it should handle way more than 
> 3cps - if I made a build of the repro proxy for windows, could you do 
> any performance testing on it and see how that goes?
>
>
>  On 5/16/05 4:43 AM, "Moshe Matitya" <moshe.matitya at kayote.com> wrote:
>
>> I am running the StatelessProxy application, which is included in the 
>> reSIProcate distribution, on Windows XP SP2 with a Pentium 4 
>> processor.  It is configured to open only UDP ports.  I set up a load 
>> test in which every call session consists of the following four 
>> messages:  INVITE, 180, 200, ACK.  The scenario is as follows:
>>
>>                      reSIProcate
>>                     .  Stateless  .
>>                   .      Proxy      .
>>                 .                      .
>>               .                          .
>>         Phone                             Phone
>>           A                                 B
>>           |                |                |
>>           |    INVITE F1   |    INVITE F1   |
>>           |--------------->|--------------->|
>>           | 180 Ringing F8 | 180 Ringing F8 |
>>           |<---------------|<---------------|
>>           |    200 OK F11  |    200 OK F11  |
>>           |<---------------|<---------------|
>>           |     ACK F12    |     ACK F12    |
>>           |--------------->|--------------->|
>>           |                                 |
>>
>> What I've found is that the Stateless Proxy cannot handle any more 
>> than 3 calls/sec.
>>
>> I presume that I'm doing something wrong, since the reSIProcate 
>> Project page [<http://www.sipfoundry.org/reSIProcate/>] states that 
>> reSIProcate provides "fast performance: > ~500 transactions per 
>> second on an Intel(TM) P4".
>>
>> Is the StatelessProxy application supposed to handle throughput on 
>> this scale?  If it is, then what should I be doing to boost the 
>> performance up to the expected levels?
>>
>> Thanks,
>>
>> Moshe
>>
>> _______________________________________________
>>  resiprocate-devel mailing list
>>  resiprocate-devel at list.sipfoundry.org
>>  https://list.sipfoundry.org/mailman/listinfo/resiprocate-devel
>
>  _______________________________________________
> resiprocate-devel mailing list
> resiprocate-devel at list.sipfoundry.org
> https://list.sipfoundry.org/mailman/listinfo/resiprocate-devel