I'm actually more interested in the performance of Kannel as a WAP
gateway. I'm really concerned as my CPU is occupied by 99% with wapbox
... given that I'm running 3 GHz Xeon and only have a couple of test
users!

A simple 'strace' shows contineous 'poll'ing.

I'm not sure how normal is that.

On Fri, 6 Aug 2004 12:48:04 -0300, Rodrigo A. Cremaschi
<[EMAIL PROTECTED]> wrote:
> 
> Hi Yaman,
> 
>     I'm using fakesmsc to stress test Kannel + server and evaluate their
> performance. Of course the results depend a lot on the server configuration,
> number of processors, etc. If you run fakesmsc in a different machine or if
> you access a database to get the answers sent back to fakesmsc, also affects
> performance.
> 
>     These tests do NOT interact with the SMSCs and are only useful to test
> your particular installation/configuration/server/application/database.
> 
>     If you need any, I can provide some configuration/results which may not
> be very prepresentative, because they are "lab tests", not real life ones.
> 
>     Regards, Rodrigo.
> --
> 
> 
> ----- Original Message -----
> From: "Yaman Saqqa" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, August 03, 2004 3:38 AM
> Subject: Performance
> 
> > Guys .. anyone knows of any stress test fugures regarding kannel's wap
> > gateway and/or any lmitations on maximum number of concurrent
> > connections ... etc
> >
> > Thanks much,
> >
> > --
> > abulyomon
> >
> > KiLLTHeUPLiNK
> 
> 
> Este mensaje se dirige exclusivamente a su destinatario y puede contener
> informaci�n CONFIDENCIAL sometida a secreto profesional o cuya divulgaci�n
> est� prohibida en virtud de la legislaci�n vigente. Si ha recibido este
> mensaje por error, le rogamos que nos lo comunique inmediatamente por esta
> misma v�a o por tel�fono (54.11 5776-5000) y proceda a su destrucci�n.
> N�tese que el correo electr�nico v�a Internet no permite asegurar ni la
> confidencialidad de los mensajes que se transmiten ni la correcta recepci�n
> de los mismos. En caso de que el destinatario de este mensaje no
> consintiera la utilizaci�n de correo electr�nico v�a Internet rogamos lo
> ponga en nuestro conocimiento de manera inmediata.
> 
> 


-- 
abulyomon

KiLLTHeUPLiNK

Reply via email to