Re: VCC cost models .... (was Re: Limits on SVCCs)

["Eric W. Gray" <[e-mail]>]

Grenville Armitage wrote:
> 
> >>The recent discussion has been suggesting that a cut-through would
> >>"cost more" than a routed path across an NBMA fabric.
> >>
> >>Maybe I am looking at this differently, but it seems to me definitional
> >>that a cut-through costs the same or less in real terms.
>         [..]
> >>There are many possible charging models for
> >>IP service, and many possible charging models for lower layer service.
> 
> There are possible models, and then there are existing models. I'm
> not aware of any ISPs serving 'the masses' who charge users for
> how many router hops their outbound traffic uses. I am aware of
> lower link providers who charge based on the topological 'distance' of
> the calls. Given such real models, minimizing router hops by
> making a long distance call comes across as less than obviously
> beneficial (financially) to the entity that was redirected to perform
> the cut-through. Sure, it saves router hops, but if I as Joe User
> can get my web pages either way, I dont care for the incremental charge
> of this cut-through.
>

"... less than obviously beneficial (financially)" - hmmm, sounds good.  Studio
audience says - BBBRRRRTT!!

Shortcuts are intended to be created to handle high volume flows.  The numbers I
have heard kicked around are the time honored 80-20 (80% of the traffic in about
20%  of the flows) which translates to a substantial reduction in the traffic level
which must be supported by the ISP.  I find it hard to believe that your ISP will
not eventually find a way to charge you for using 5 times as much bandwidth as the
other users of the ISP.

> 
> If anyone's done a deeper analysis already, and can release it,
> perhaps we can start a FAQ on cut-through. I'd be happy to see
> a more detailed argument refute this straw-man negative scenario
> constructed above (but based on existing or deployable-in-the-time-
> frame-of-NHRP charging models, of course).
> 

Please, can we be consistent about the terminology?  I believe that "cut-through"
has a completely different meaning in a related context and the term we have been
using in its place is "shortcut".

>
> cheers,
> gja

--
Eric Gray