On Tue, Dec 8, 2009 at 9:18 AM, Behcet Sarikaya <[email protected]> wrote: > > >> > Alain wrote in his mail which I quote here: >> > >> > In our case, we have: >> > >> > * many millions of edge devices (similar to your UE) >> > * thousands of routers >> > * thousands of servers >> > >> > So your mileage with DS-Lite varies according to how much your network fits >> into the above proportions. If you exactly have the above, DS-Lite fits you >> best. >> > >> >> Nope, that does not sound like 3GPP mobile. 3GPP architecture is >> completely based on tunneling. The only devices that need IPv6 >> addresses or IPv6 routing capability are the UEs and the GGSN, and we >> have support in those devices today. > > This is what I meant, in 3GPP mobile you don't have millions of home routers > that need IPv6 addresses. > > I think we are agreeing on this.
Umm..... right, i have millions of edge devices that need addresses...aka UE... and UE can grow exponentially with machine to machine communication. Also, with the evolution of MiFi, we'll see that the mobile carrier UE may be a WiFi router just like the Cable MSO operators... Services are evolving and the differences are fading between wireline and wireless. > >> I am not sure what your conclusion is, given that your facts regarding >> base stations (do you mean BTS? NodeB?) and backhaul are not correct. > > If we can get them do IPv6, it would even be better, I think. Agreed. I believe most LTE / 4G E-UTRAN will be native IPv6, but that is greenfield. My only point being that IPv6 can come to 2G/3G mobile devices without any major technical barriers today. > > >> I frankly don't think DS-lite works well for mobile, > > I don't want to put it this way, as I said it is easy to deploy DS-Lite in > mobile, maybe except for AFTR box. > > I won't say DS-lite is easy or not, but i will say native IPv6 or native dual stack on the UE is easier than DS-Lite. > Regards, > > Behcet > > > > _______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
