[28712] in North American Network Operators' Group
Multi-home II
daemon@ATHENA.MIT.EDU (Rural CNE)
Sun May 14 01:09:04 2000
Message-ID: <007d01bfbd62$5d27d740$0200a8c0@mulekick.net>
Reply-To: "Rural CNE" <bwalters@inet-direct.com>
From: "Rural CNE" <bwalters@inet-direct.com>
To: <nanog@nanog.org>
Date: Sun, 14 May 2000 00:07:17 -0500
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Errors-To: owner-nanog-outgoing@merit.edu
> (There are very few valid reasons to multi-home, please consult
> with RFC x, BCP y, and an experienced network engineer.)
> Roger that.
Some of the reasons I have seen discussed have included:
1.) Reliability / Availability
2.) Geographically distributed client locations
I would like to add:
3.) Performance / Throughput
The old "we'll just add another subinterface to the hssi port, they
can't tell the difference anyway" school of network design just ain't
gonna cut it anymore. The Internet has now become the e-Internet.
Upstream providers are not created equal. IAP's and other
end-systems will be migrating to, or multi-homing with, those service
providers that can handle the additional traffic load generated by
xDSL, cable, or other high bandwidth technology.
(...And your point is?)
I would argue that lower tier multi-homing has contributed to the
growth of the CIDR tables. Examining and resolving the issues
surrounding the cause, multi-homing, would be more beneficial than
promoting an upstream provider policy dealing with the effect,
CIDR growth.
-brad (Rural CNE)
bwalters@inet-direct.com
