[Asterisk-Users] Concurrents calls on asterisk with H323

[John Todd]

>Look for the recent 'capacity testing' thread here. We've had some 
>discussions on it, but so far the bottom line sounds like you won't 
>be able to run more than 20 - 25 decent quality calls before 
>asterisk dies.
>
>jesse
[snip]

Your statement relies completely on assumptions which may be 
incorrect.  Transcoding significantly degrades performance, but 
without transcoding it may be possible for * to move dozens or 
hundreds of calls with H.323.  See note below.

JT




>Subject: RE: [Asterisk-Users] A question on codec translation.
>From: "Tom Lowe" <tom at comprotech.com>
>To: <asterisk-users at lists.digium.com>
>Reply-To: asterisk-users at lists.digium.com
>Date: Mon, 12 Jan 2004 08:45:21 -0500
>
>If the incoming and outgoing Codecs are the same, there is no
>"conversion" done.  It basically becomes a packet relay, what goes in,
>comes out.
>
>I'm not sure of the answer to your second question.  However, your
>question actually begs a question I've been wondering about in the last
>couple of days:
>
>I'm doing H.323 in, H.323 out....simple relay.  (This is my customer's
>requirement...not my preference).  What I want to do is ALWAYS use the
>same codec for the outgoing leg as for the incoming leg.  In other
>words, if the call comes in as G.729, the outgoing call uses G.729 ONLY.
>If the incoming call is G.711, I want the outgoing to be G.711.   I want
>to avoid any sort of transcoding.
>
>Is it possible?
>
>Thanks.
>
>Tom Lowe
>
>(FYI, Dual Xeon 3.06, 120 channels (60 calls) of above scenario, G.729
>using less than 10% CPU!)  (Remember, no transcoding is being performed)
>