[5625] in North American Network Operators' Group
Re: Ungodly packet loss rates
daemon@ATHENA.MIT.EDU (Alex.Bligh)
Tue Oct 22 20:18:08 1996
To: jon@branch.net (Jon Zeeff)
cc: jcurran@bbnplanet.com (John Curran), kwe@6sigmanets.com, nanog@merit.edu
In-reply-to: Your message of "Tue, 22 Oct 1996 18:32:49 EDT."
<m0vFpNS-000H5wC@aero.branch.com>
Date: Wed, 23 Oct 1996 00:29:00 +0100
From: "Alex.Bligh" <amb@xara.net>
>
> > While there is no difference from a technology perspective,
> > there's also no benefit to be gained by interconnecting large
> > networks at a public (as opposed to private) interconnects.
>
> Sure there is. Running one larger interconnect has the potential to
> be much less expensive and easier to run than many smaller ones.
> Economy of scale...
>
For whom? Just as a matter of curiosity, lets say 5 NSPs
happen to share colo/FM at various locations across the US.
What is the incremental cost to them of a back to back HSSI/FDDI
connection? Almost certainly smaller than the lower performance
and greater adminstrative hassle of paying someone else in the
same facility running a NAP to move the packets for you,
which also entails greater risk and unreliability. I would
guess viability of the NAP model vs. the private interconnect
model depends to a great extent on (a) peering policy and
(b) for any given peer, what the traffic between networks is.
Alex Bligh
Xara Networks
