Re: [multimob] IGMP/MLD Proxy at the MAG

["Thomas C. Schmidt" <[e-mail]>]

Hi Luis,

Luis M. Contreras wrote:
> Hitoshi Asaeda <[e-mail]> dijo:
> 
>>
>> MLD modification on a host-side implementation is not needed.
>> I've been saying, since a MN does not give its join state to its
>> upstream router (incl. proxy) unless no query reception, the MN cannot
>> continue to reveive streams from nMAG until nMAG completes joins
>> for all the subscribed channels.
>> We are trying to tune some MLD timer value to *reduce* this join
>> latency (a few seconds or more?) without protocol modification at this
>> stage, but it does not *eliminate* or effectively reduce the latency.
>>
> 
> I think the method we propose in draft-contreras-multimob-msd-00 to 
> support the MN handover by temporally redirecting the current multicast 
> flow (subscribed by the MN) from pMAG to nMAG till (normal or tuned) 
> subscription renewal can effectively reduce the latency in the reception 
> of the multicast content once the MN is attached to nMAG, at the cost of 
> more bandwidth temporally in the core network, as well as some other 
> inefficiencies (temporal routing ineficiency, MTU reduction, etc). I 
> think it is something we have to trade off.
> 

You do point at the proposal redirecting via the LMA?
Mhmm, then are you sure about this? We discussed the temporal handover 
behavior prior to IETF 76 online, and if I remember correctly, timing 
was exactly the same as for the "full" handover ??

Cheers,

Thomas


-- 

Prof. Dr. Thomas C. Schmidt
° Hamburg University of Applied Sciences                   Berliner Tor 7 °
° Dept. Informatik, Internet Technologies Group    20099 Hamburg, Germany °
° http://www.haw-hamburg.de/inet                   Fon: [phone] °
° http://www.informatik.haw-hamburg.de/~schmidt    Fax: [phone] °