Date: Sun, 5 May 1996 13:52:32 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: JimFleming@unety.net (Jim Fleming) Cc: avalon@coombs.anu.edu.au, imb@scgt.oz.au, FreeBSD-hackers@FreeBSD.ORG, JimFleming@unety.net Subject: Re: IPv8 Tutorial #1: Minimal IPv8 hack Message-ID: <199605052052.NAA20010@phaeton.artisoft.com> In-Reply-To: <01BB3A2C.4CFF6A80@webster.unety.net> from "Jim Fleming" at May 5, 96 02:41:01 am
next in thread | previous in thread | raw e-mail | index | archive | help
Jim Fleming writes:
> On Sunday, May 05, 1996 12:33 PM, michael butler[SMTP:imb@scgt.oz.au] wrote:
> @ Darren Reed writes:
> <snip>
> @ Neither IPv8 or IPv6 is going to magically "fix" the result of at least
> @ one of the three 6 meg bearers taking today off on a picnic :-(
> @
> @ michael
> @
> @
>
> Keep in mind that if you are on an IPv8 network then the
> IPv4 Legacy Internet is viewed as "damage" and we route
> around it...:-)
You are dragging the *worst* attribute of IPv4 and IPv6 with you
into "IPv8": connection to servers instead of connection to
services.
The way to resolve the route congestion problem from your other
posting is not bigger pipes (such as might be temporarily provided
through alternate routes, until they, too, filled up), it's less
use of the congested route ("congested route" in the "IPv8" scheme
would refer to multiple route congestion, but it's topologically
equivalent). To accomplish this, you need to replicate services
on your side of a congested link, and connect to the service,
instead of the server, to avoid use of the link altogether.
This does not seem a direct fallout of your "Limited Hierarchical
Routing".
I am also more than somewhat concerned with your geographical
"galaxy" assignment, and the fact that the increased address
space may buy us some time, but will not buy as much time as IPv6.
I suggest that FreeBSD adopt equivalent code that will not hinder
your research, when choosing between one set of code or another,
but also that it should not integrate the "IPv8" code directly,
especially with the "hacks" you've suggested. This course of
action on the part of FreeBSD is simply optimization of choices
to increase available options, and is philosophically unrelated
to support for or condemnation of "IPv8".
This *can* be seen as an opportunity for better functional code
organization, such that you could load your "IPv8" code into an
already booted kernel to support "IPv8", but I think the onus for
the interface code is squarely on your shoulders, not on wholesale
integration of "IPv8" into FreeBSD.
Regards,
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605052052.NAA20010>