[93785] in North American Network Operators' Group
Re: Home media servers, AUPs, and upstream bandwidth utilization.
daemon@ATHENA.MIT.EDU (Frank Coluccio)
Mon Dec 25 20:58:52 2006
From: Frank Coluccio <frank@dticonsulting.com>
To: Lionel Elie Mamane <lionel@mamane.lu>,
Simon Leinen <simon@limmat.switch.ch>
Reply-To: frank@dticonsulting.com
Date: Mon, 25 Dec 2006 19:57:52 -0600
Cc: NANOG <nanog@merit.edu>
Errors-To: owner-nanog@merit.edu
>doesn't save much. A possible countermeasure is to not count off-peak
>traffic (or not as much). Our charging scheme works like that, but
>our customers are mostly large campus networks,
This is similar to what some schools of economic study suggest in order to
achieve equilibrium while attempting to rationalize theoretical models that=
are
placed under stress. The problem here, of course, is that economists workin=
g on
theoretical models don't have to concern themselves with physics, whereas f=
irst
mile network providers do.=20
Instead of devising variations of schemes that are based on traffic averagi=
ng- or
95 percentiles (designed primarily for colocation- and IX Point situations),
access network providers must address customer facing bandwidth utilization=
and
provisioning issues, and while the issues of upstream transiting and peerin=
g are
related, they nonetheless are exogenous to the access network problem.=20
The main problem areas affecting the original poster's concerns over the ab=
ility
to send traffic in the upstream direction, as I see them:=20
[a] the nature and being of telcos' DSL and MSOs' cable modem asymmetrical
network designs, and,=20
[b] the unrealistic expectations that most stakeholders have (or the bill of
goods that have been successfully sold by incumbents to the community at la=
rge)
that suggests that an equitable exchange of money for services could be obt=
ained
from a model that depends on entirely statistical probabilities.
The ultimate answer to this problem is a lot of discussions and subsequent
actions that will need to be taken. However, at the level of line granulari=
ty
that exists at the CO hub or head end terminal gear, the 95 percentile pric=
ing
scheme is not relevant.
Frank A. Coluccio
DTI Consulting Inc.
212-587-8150 Office
347-526-6788 Mobile
On Mon Dec 25 16:50 , Simon Leinen sent:
>
>Lionel Elie Mamane writes:
>> On Mon, Dec 25, 2006 at 12:44:37AM +0000, Jeroen Massar wrote:
>>> That said ISP's should simply have a package saying "50GiB/month
>>> costs XX euros, 100GiB/month costs double" etc. As that covers what
>>> their transits are charging them, nothing more, nothing less.
>
>> I thought IP transit was mostly paid by "95% percentile highest speed
>> over 5 minutes" or something like that these days? Meaning that ISP's
>> costs are maximised if everyone maxes our their line for the same 6%
>> of the time over the month (even if they don't do anything the rest of
>> the time), and minimised if the usage pattern were nicely spread out?
>
>Yes. With Jeroen's suggestion, there's a risk that power-users'
>consumption will only be reduced for off-peak hours, and then the ISP
>doesn't save much. A possible countermeasure is to not count off-peak
>traffic (or not as much). Our charging scheme works like that, but
>our customers are mostly large campus networks, and I don't know how
>digestible this would be to retail ISP consumers.
>--=20
>Simon.
