[asterisk-dev] asterisk-dev Digest, Vol 51, Issue 50

[Di-Shi Sun]

Hi Maxim,

We used SIPp to send  audio files to generate the RTP traffic. As I
understanding, it were 214 (frame size) /172 (UDP size) byte for g711 and 74
(frame size) / 32 (UDP size) byte for g729 RTP packets in the test. If you
are interested in the audio files, we can sent them to you. They are too big
to be sent to the maillist.

Regards,

Di-Shi Sun
VoIP Routing, Accounting, Security
www.TransNexus.com

>
> Message: 4
> Date: Thu, 23 Oct 2008 02:35:11 -0700
> From: Maxim Sobolev <sobomax at sippysoft.com>
> Subject: Re: [asterisk-dev] asterisk with osp performance test results
> To: Asterisk Developers Mailing List <asterisk-dev at lists.digium.com>
> Message-ID: <4900454F.5030506 at sippysoft.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> John Todd wrote:
> > I would not expect NAT to create any significant decrease in
> > performance on these results, since the packet re-write on the RTP
> > packets is fairly lightweight from what I understand.
>
> The RTP doesn't need any rewrite in NAT scenario vs. no-NAT scenario.
> SIP does, but it's unlikely to affect simultaneous calls count
considerably.
>
> The test itself is not complete, though, as the performance is likely to
> depend heavily on the RTP PPS (packets per second) rate, which could
> range from 100 to 33 depending on the frame size in use. I also could
> not find the frame size in the report, though it's possible that I have
> overlooked it.
>
> Regards,
> -- 
> Maksym Sobolyev
> Sippy Software, Inc.
> Internet Telephony (VoIP) Experts
> T/F: +1-646-651-1110
> Web: http://www.sippysoft.com
>