peering charges?

Matt Harrop mharrop at interlog.com
Sun Jan 26 10:52:16 UTC 1997


At 02:38 AM 1/26/97 -0800, you wrote:
>Jonathan Heiliger <loco at isi.net> wrote:
>
>>Large ISPs should probably be interested in access to content, without it
>>their users could find the Internet a very boring place.
>
>Not only that.  They are also interested in peer's customer population
>gaining access to content (and ads) published by their own customers.
>
<My point of view may be very different from most of the list since I'm in
Toronto, Canada>

  From my point of view (local ISP) I simply look at the cost versus
performance.  We're connected to our primary transit provider (UUNet
Canada) via 100MB fibre, and we pay for average bandwidth usage.  If paying
for a non-transit (peering only) T1 into Sprint Canada ends up giving me
equal or better performance for less money than paying for another 1.2Mbit
of usage on my UUnet connection I'll be happy to pay.

  Whether I should be paying Sprint Canada for peering, they should be
paying me, or we shouldn't charge each other a dime isn't relavent.  When
it comes right down to it either I'll peer or I won't based on the
economics...  If they want to charge an unreasonable sum (defined as more
than it would cost me to send the traffic through UUnet onto Sprint Canada)
then I won't peer with them.  In which case they lose the revenue that I
might have been willing to provide them.  I'm still happy because I'm
getting the most bandwidth for the least dollars.

My case may well be fairly unusual in that we have big bandwidth (for a
local ISP) but service only one area code, so we don't have to fund any
long haul network of our own.  Just our luck to be in the single largest
(by population) local calling area on the planet ;)

My $0.02.







More information about the NANOG mailing list