Re: [Mip4] RE: Differences between Low Latency Handovers and FMIP v4

["James Kempf" <kempf at docomolabs-usa.com>]

Hi Rajeev,

> >Unfortunately, vendors would like to keep it that way, since having a
> >standardized solution means that enterprise customers and service
providers
> >would be free to mix and match solutions and less likely to go single
> >supplier. So FMIP is likely to be of limited interest for WLAN
deployments
> >going forward. Perhaps 3GPP2 is interested in it? Though they already
have
> >
> >
> On the contrary, wouldn't you expect the enterprises to start demanding
> interoperability
> between boxes? If vendors have "holds" on enterprises for their boxes
> alone, that does not
> sound like a way going forward.
>

Well, one would hope.

> >Anyway, if this WG was serious about its charter to focus on deployment,
it
> >would start up an activity to standardize WLAN switching "IP mobility".
> >
> >
> >
> Doesn't this mean tunneling?  A switch tunneling an L2 frame in L3
> sounds sub-optimal
> since it needs to "go back" through the router to reach the IP network,
> and eventually the
> destination router and the switch. Anyway, why is there a need for
> standardization here?
>

There's a bit  more of a dogleg on the routing than on something like FMIP.
With FMIP, the dogleg is from the crossover aggregation router between the
previous and new ARs and the previous and new ARs. With the WLAN switches,
there's that and between the switch and previous AR.

As for what's to standardize, essentially, these protocols work like
Postreg. So there would need to be some agreement about the inter-switch
signaling to set up the tunnel, and the actual tunnel type. I don't know
offhand what kind of tunnel is used, I suspect each switch might use a
different tunnel type.

            jak



-- 
Mip4 mailing list: Mip4 at ietf.org
    Web interface: https://www1.ietf.org/mailman/listinfo/mip4
     Charter page: http://www.ietf.org/html.charters/mip4-charter.html
Supplemental site: http://www.mip4.org/