[123768] in North American Network Operators' Group
RE: Need some info about "Clean pipe"
daemon@ATHENA.MIT.EDU (Brandon Kim)
Mon Mar 15 14:58:59 2010
From: Brandon Kim <brandon.kim@brandontek.com>
To: <nanog@nanog.org>
Date: Mon, 15 Mar 2010 14:58:32 -0400
In-Reply-To: <BA8F7071-2364-4DD0-B74F-ABA682BF1C23@arbor.net>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Is this a new concept? I've never heard of this before. It's very interesti=
ng. Not that I personally have
a need for it=2C but companies are always finding more "services" to provid=
e for you....errr....manage for you.....
> From: rdobbins@arbor.net
> To: nanog@nanog.org
> Date: Mon=2C 15 Mar 2010 18:35:29 +0000
> Subject: Re: Need some info about "Clean pipe"
>=20
>=20
> On Mar 16=2C 2010=2C at 1:06 AM=2C Michael Holstein wrote:
>=20
> > In short=2C instead of paying for a (n*)gbps circuit and buying your ow=
n DDOS prevention gear=2C you buy $n worth of bandwidth that has somebody a=
ctively managing the DDOS protection.
>=20
> And of course=2C if one's organization is an SP=2C one can in fact offer =
this type of service commercially to one's transit/hosting/co-location/ASP/=
cloud/etc. customers.
>=20
> =3B>
>=20
> Responding to the original poster's question about latency=2C if the serv=
ice architecture is well-defined and takes backhaul-induced latency into ac=
count as part of the design/topological service coverage=2C latency experie=
nced by the end-customer is typically minimal.
>=20
> -----------------------------------------------------------------------
> Roland Dobbins <rdobbins@arbor.net> // <http://www.arbornetworks.com>
>=20
> Injustice is relatively easy to bear=3B what stings is justice.
>=20
> -- H.L. Mencken
>=20
>=20
>=20
>=20
=
