[arin-ppml] IPv6 /32 minimum for extra-small ISP

[David Farmer]

Milton L Mueller wrote:
> How would one get a bead on how many do and do not take a full table and how that ratio has changed over time? 

<snarky on>

At the risk of creating an matter-antimatter reaction within ARIN's list 
server, I will use both Randy's and Milton's names in the same email.  I 
will apologize now if the chain reaction destroys the universe.  :)

<snarky off>

Randy Bush and others have the done some related work;

http://psg.com/~olaf/measurements/as3130/measurements.html

For paper and slides;

http://psg.com/~olaf/measurements/as3130/publications.html

Quick summary from there data;

		# tested 	default 	default-free 	mixed
stub 		24,224 		77.1% 		19.3% 		3.6%
small ISP 	1,307 		44.5% 		42.2% 		13.3%
large ISP 	246 		17.1% 		60.6% 		22.3%

You could probably come up with some kind of model to create an estimate 
of how many routers that represents.

When I look at this data it seem to me that it supports the general 
hypothesis those that add prefixes to the DFZ are probably not sharing 
equally in the costs associated with the additional prefixes.

Whoops, there she Blow......

:)

-- 
===============================================
David Farmer               Email:farmer at umn.edu
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota	
2218 University Ave SE	    Phone: 612-626-0815
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================