[109558] in North American Network Operators' Group
Re: an over-the-top data center
daemon@ATHENA.MIT.EDU (Patrick W. Gilmore)
Mon Dec 1 11:54:06 2008
From: "Patrick W. Gilmore" <patrick@ianai.net>
To: NANOG list <nanog@nanog.org>
In-Reply-To: <1EF5A80BBAAC8DA3299E017A@rasmus.kthnoc.net>
Date: Mon, 1 Dec 2008 11:53:58 -0500
Errors-To: nanog-bounces@nanog.org
On Dec 1, 2008, at 11:06 AM, M=E5ns Nilsson wrote:
>> End of day, an IXP is not some magical thing. It is an ethernet =20
>> switch
>> allowing multiple networks to exchange traffic more easily than =20
>> direct
>> interconnection - and that is all it should be. It should not be =20
>> mission
>> critical. Treating it as such raises the cost, and therefore =20
>> barrier to
>> entry, which lowers its value.
> You did not answer my question on usability of fiber based on amount =20=
> of
> knowledge about where it is.
Of course knowing where the fiber is does not stop the backhoes. It =20
was obvious you were being silly, so I ignored it. By that logic, =20
providers should not check any fiber path they purchase because it =20
will not stop the backhoes. I suspect most providers will continue to =20=
buy from multiple providers, check the paths themselves, ensure =20
grooming onto a single path is not a problem, and several other =20
perfectly valid operational best practices which are impossible at =20
NetNod.
OTOH: My paragraph above yours is a serious consideration, which you =20
have blithely ignored.
As I said before, feel free to use what you please, where you please. =20=
Your network, your decision. I frequently do things which would not =20
be considered best practices in certain instances, but that does not =20
make them valid for everyone everywhere, and I would not argue such.
--=20
TTFN,
patrick
