Re: How long will it take to completely get rid of IPv4 or will it happen at all?

[Scott Morizot <tmorizot () gmail com>]

On Sat, Jun 27, 2015 at 9:38 AM, Bob Evans <bob () fiberinternetcenter com>
wrote:

> What will come first ?
> A) the earths future core rotation changes altering the ionosphere in such
> a way that we are all exposed to continuous x-rays that shorten our
> lifespan
>                  OR
> B) the last IPv4 computer running will be reconfigured to IPv6
>
> Thank You
> Bob Evans
> CTO

At least from a large enterprise perspective, I don't really care when IPv4
is removed from that last computer. Instead, I care about how long it will
take us to eliminate IPv4 from most or all of our internal network and
confine its continued support to our dual-stacked public resources and
legacy support at our perimeter. In particular, our plans right now focus
on transitioning to a native IPv6-only wide area network providing legacy
protocol support where needed using LISP. (We already have LISP configured
and deployed to our largest sites.) We're in the process of ensuring all
clients are dual-stacked and deploying IPv6 to internal applications. We
are testing and developing a process to create IPv4 "enclaves" in our data
centers for applications that cannot timely transition fronted by NAT64 so
we can start removing IPv4 from our many smaller access network sites.

It's not really our problem or concern how long some people choose to keep
IPv4-only systems running, even as those systems increasingly become
second-class citizens on the network. Running a large, fully dual-stacked
enterprise network is overly-complex, increases costs, and imposes
limitations. As time proceeds, I expect most enterprises that haven't
already done so will reach a similar conclusion.

I've never worked at a carrier or ISP, so I have no particular insight into
the drivers pushing those sorts of networks. But the presentation by
Comcast on possible plans to provide long term legacy IPv4 support as an
overlay service suggest to me that the drivers are not completely
dissimilar from their perspective.

So it really doesn't matter that much how long IPv4 continues to exist in
one sense or another. It's the tipping point where much of the Internet
begins to treat it as a second-class citizen that really matters. I would
suggest most people will not like ending up on the wrong side of that curve.

My perspective, anyway.

Scott