Re: [v6ops] University WiFi (was: Implementation Status of PREF64)

[David Farmer <[e-mail]>]

On Tue, Oct 12, 2021 at 12:44 PM Gert Doering <[e-mail]> wrote:

> Standard univerisity WiFi deployment today seems to be "RFC1918 space
> and a huge NAT box"...
>

I can tell you why we do this; We have light rail tracks and several major
roads that go through our campus. As a result, several hundred to
thousands, some hours many thousands, of devices, attach to our public WiFI
every hour and never do anything other than getting a DHCPv4 address from
us and maybe probing an aliveness sight of their choice. These devices
usually come and go in a matter of minutes. We would either need
unreasonably short DHCP lease times or use large RFC1918 pools so we don't
have to worry about recovering addresses very quickly.

We intentionally have good WiFi coverage around the three train stations on
campus and for the many bus stops and sidewalks across campus to provide
coverage while people wait for public transportation.

Even for our 802.1x network, we have a very large number of users, +50,000
and they move around campus significantly, and between our mobility
addressing domains, consumption of address varies significantly across the
day in the various parts of campus, up to 10s of thousands in
hourly transitions. For IPv6 addresses, these consumption patterns really
aren’t an issue. However, for IPv4 if we used globally unique addresses we
would need several times the overall usage to compensate for the variation
in consumption patterns.

Hope that helps.

-- 
===============================================
David Farmer               Email:[e-mail]
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota
2218 University Ave SE        Phone: [phone]
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================