Re: [v6ops] LISP support for draft-ietf-v6ops-rfc7084-bis-00

["STARK, BARBARA H" <[e-mail]>]

> Fully agree with Cameron view, and furthermore … if we need to choose one
> mechanism because how many millions of devices already use it, and this
> demonstrates the success, clearly 464XLAT has more TENS OF MILLIONS of
> users than DS-LITE + 6RD + MAP E + MAP T, altogether, and of course it
> means a single ISP having both wired and cellular networks, can have a single
> way to deploy IPv6 instead of two.

RFC 7084 has nothing to do with cellular (3GPP) networks. 3GPP deployments are not relevant to this discussion. 3GPP and wireline deployments are not currently converged, with the 3GPP RAN and IP core networks being distinct from the wireline IP access networks (wireline is used as backhaul for the 3GPP networks at layers 1 and 2, but not for IP). What makes sense for a 3GPP deployment may or may not make sense for a wireline deployment -- especially considering the lack of extensive IA_PD support in deployed 3GPP networks.

If some ISP with wireline and wireless has interest in converging their wireline IPv6 deployment with their wireless, in the very near future (using network elements that are already being trialed or have been deployed), it would indeed be nice for them to speak up. Personally, I have no internal requests from my company to drive such IPv6 convergence.

So please don't use statistics related to 3GPP deployments in an argument for wireline CE routers. And please don't try to bloat the CE router with everything needed to support attachment to a 3GPP network.

In the wireline deployments I'm aware of, the only ones reaching into the millions have used dual stack, 6rd, or (possibly) DS-Lite. And of these 3, the only one that I believe should ever be considered mandatory for all CE routers is dual stack.
Barbara