owen at DeLong.SJ.CA.US
Sat Aug 22 18:54:36 UTC 1998
> If we look at any other downstream, or end, customer, they still pay their
> provider for a full circuit's worth of bandwidth. Even if they are not
> symmetrically using their bandwidth. So, you pay not only the telco, but
> your upstream provider for use of data in both directions. ISPs use
> customers like this to not only balance their traffic patterns with other
> services that suck massive bandwidth (such as universities and dialup), but
> to provide additional revenue (by not discounting a circuit in half because
> they primarily use traffic headed in one direction)
OK, I completely agree with you so far.
> A network that has chosen a business model that is tailored towards outbound
> only transit should be prepared to pay for transit the same way their
> customers do. Their customers should pay a premium to have a network that
> is better connected than they could provide themselves, not so their
> upstream can reap the benefits of not having to pay for their own transit.
> These companies have opted towards a business model that does not balance
> traffic, thus making it difficult to be a "peer" with a network.
I don't get the connection. True, Exodus has a business model tailored towards
OUTBOUND flows, but not OUTBOUND TRANSIT. The outbound flows are exactly a
portion of the INBOUND traffic above. Nothing else.
> This, of course, brings up many, many interesting questions about which
> companies have the right to be at the top of the network pyramid. Although
> it seems the market place has already determined this, at least a majority
> of it.
That's assuming a pyramid is necessary, desirable, etc. In fact, there's
no real need for such a pyramid, but at the moment, there does seem to be
a small number of larger providers with increasing numbers of providers as
you reduce the size of said provider.
> This shift in costs have been a long time coming, you can't charge your
> customers for bandwidth, and then not expect to have to pay for it yourself.
Exactly. That's exactly why I think that Exodus/BBN are peers and that the
assymetrical nature of the relationship is mostly irrelevant at this point.
As you've pointed out, the BBN customer pays BBN for a full circuits worth
of bandwidth. Exodus provides to BBN what those BBN customers request from
Exodus. Exodus does not charge those BBN customers (or BBN) for the requests or
related ACKs, nor does Exodus charge the BBN customers or BBN for the data
being delivered across Exodus network to the point where BBN has told Exodus
they would like to receive the traffic for that customer. Given that, I don't
see how you say anyone is currently charging customers for bandwidht and not
paying for it themseleves. Exodus runs a very large national network, with
very high growth rates.
> jamie at networked.org
> The views stated above are mine and do not reflect those
> of my employer.
The views I have expressed are mine and may or may not coincide with my employer.
> -----Original Message-----
> From: Michael Dillon [mailto:michael at memra.com]
> Sent: Friday, August 21, 1998 7:44 PM
> To: nanog at merit.edu
> Subject: Re: BBN/GTEI
> On Fri, 21 Aug 1998, Karl Denninger wrote:
> > No, its actually becoming MORE suitable. Instead of burning the entire
> > circuit in both directions, you're only burning half of it now (one
> > direction).
> You still have to pay for the whole circuit.
> Michael Dillon - Internet & ISP Consulting
> Memra Communications Inc. - E-mail: michael at memra.com
> Check the website for my Internet World articles - http://www.memra.com
More information about the NANOG