[arin-ppml] ARIN-prop-127: Shared Transition Space for IPv4 Address Extension

Mark Smith ipng at 69706e6720323030352d30312d31340a.nosense.org
Sat Jan 22 18:55:33 EST 2011

On Sat, 22 Jan 2011 16:16:36 -0700
Chris Grundemann <cgrundemann at gmail.com> wrote:

> On Sat, Jan 22, 2011 at 15:22, Mark Smith may have written:
> > more likely absolute maximum of 128K, and probably more
> > commonly no more than 32K.
> For a little insight into how much of the industry is thinking about
> deploying LSN/CGN, you may want to take a look at a couple of the
> bigger vendors current pitches (other vendors have similar stories but
> these where the most recent two I have read):
> https://www.cisco.com/en/US/prod/collateral/iosswrel/ps6537/ps6553/brochure_c02-560497_ns1017_Networking_Solutions_Brochure.html

Well aware of that one. No use to anybody who can't afford a CRS1. Not
a good place to put this function anyway, presuming you're using
CRS1 as core or distribution routers - the best way to scale network
functions is to perform it as close as possible to the edge. Even Cisco
agree with that - remember the "no ACLs in the core, put them at the
distribution or access layer" model? The only rational reason I can
think of building one of those blades for a CRS1 is to prove that you
can do it, and that scaling something down (i.e building a smaller one)
is usually a significantly easier task than scaling something up. I've
been expecting either the code or the hardware to be released for lower
end platforms such as the ASRs. It hasn't happened yet, as far as I can
tell, based on the recent release manuals. Renaming traditional NAPT
functionality "carrier grade" doesn't make it so - mechanisms such as
recording who was using what address and port range at a particular
time for law enforcement purposes are what makes a NAPT solution
"carrier grade".

> http://www.juniper.net/us/en/local/pdf/implementation-guides/8010076-en.pdf
> A snippet from the Cisco site: "1+ million connections setup per
> second for stateful NAT44."
> Food for thought,
> ~Chris
> > Regards,
> > mark.
> > _______________________________________________
> > PPML
> > You are receiving this message because you are subscribed to
> > the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> > Unsubscribe or manage your mailing list subscription at:
> > http://lists.arin.net/mailman/listinfo/arin-ppml
> > Please contact info at arin.net if you experience any issues.
> >
> -- 
> @ChrisGrundemann
> weblog.chrisgrundemann.com
> www.burningwiththebush.com
> www.theIPv6experts.net
> www.coisoc.org

More information about the ARIN-PPML mailing list