[AusNOG] IPv6 excuses

Paul Brooks paul.brooks at tridentsc.com.au
Fri May 27 14:29:21 EST 2016


On 27/05/2016 1:56 PM, Mark Andrews wrote:
>
> This has not been true for 20 years now.  The moment we were forced
> into using NAT to connect people could connect to everything they
> wanted to.  Just because we have put up with degraded service through
> neccesity doesn't mean that there isn't a issue.  CGNAT just made
> the probem worse as many workarounds don't work with CGNAT.
>
> NAT and CGNAT are stop gap mechanisms.  People have forgotten this
> as they have had to live with it for too long.

The difference is that NAT was deployed with the cost going to the end customer, in
whatever slight increase was incurred by the NAT-enabling software in the customer's
router - so there was little incentive for the service provider to do anything to
negate it. And software applications developed rendezvous functionality that provided
the illusion of any-to-any connectivity, so customers have lived with it blissfully
unaware of the complications.

CGNAT on the other hand is a cost in expensive hardware, management and now increased
Data Retention logging that is incurred primarily by the service provider - and the
cost:benefit tradeoff is against the rapidly increasing cost of acquiring sufficient
IPv4 addresses to remove the need to deploy CGNAT - or the lower cost of setting up
IPv6 functionality to move sessions off IPv4 and thereby reduce the cost and rate of
growth of CGNAT session licenses.

When even the organisations that deploy CGNAT start to see that they can also deploy
IPv6 and see their CGNAT license cost reduce by half, I remain hopeful they'll see the
business case that wasn't there earlier.

Paul.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: paul_brooks.vcf
Type: text/x-vcard
Size: 303 bytes
Desc: not available
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20160527/d8e4a2c7/attachment.vcf>


More information about the AusNOG mailing list