I believe the load test documents state that the load tests were two
test machines connected via a gigabit switch.
In this sense there was no 'internet line' chokepoint: It was just
machine A (server) and machine B (simulating many many listeners) both
plugged into the same switch, for a straight through 1gig connection
from machine A to B.
In your case, with FIOS, I think the usual ((Number of listeners *
your streaming bitrate) + 10% TCP overhead) type calculation applies,
unless Verizon is doing something else to throttle/packetshape.
Also note that if "10/0.8" is your down/up, you're limited by that
.8
up, since you'll be serving the audio from that side.
-bhance
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 4 Apr 2007 15:18:02 -0400
> From: "Chet Sapino" <csapino at twcny.rr.com>
> Subject: [Icecast] Icecast2 Server Load tests
> To: <icecast at xiph.org>
> Message-ID: <000a01c776ed$f7677df0$0401a8c0 at sapino>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello:
>
> Ed Zaleski ran some remarkable and very impressive load tests on icecast
reported on www.icecast.org (Home page). What wasn't mentioned in the test
specifications was the type of internet line that was used which totally
influences bandwidth and subesequent overall load characteristics.
>
> Questions:
>
> 1. Was it a T1 line, Verizon FIOS, or cable internet, or what?
>
> 2. What were the download/upload bandwidths in MBPS?
>
> 3. Or was this an offline simulation with no real listeners just testing
CPU load?
>
> I am using Time Warner cable internet at 10/0.8 MBPS bandwidth and am
wondering how this translates to server load streaming mp3's at 128kbps.
I.e., how many listeneers can I connect without breakup.
>
> Right now I am thinking switching to Verizon FIOS which runs at 30/5 MBPS
>
>
> TIA,
> Chet
>
