Re: [Roll] NA to transport DAO

["Pascal Thubert (pthubert)" <pthubert at cisco.com>]

Hi Richard
>
>Some things immediately jump out at me:
>     1) small packets are much more common, so DAOs may not fit into
ND.

This is a non issue. IPv6 on any link requires an MTU of at least 1280. 
6LoWPAN provides a fragmentation mechanism for those link layers that
could hold the IPv6 MTU in a frame.
Above the IPv6 MTU, you can still fragment the ND packet using IPv6
fragmentation.
This is more common than you might think, as traditional ND packets can
get real big, for instance in the presence of SeND.


>     2) multicast is not assumed, so the Whiteboard is used.

The routers could still use NS/NA between them but that makes little
sense I agree. RPL routers use NA as unicast to their parent, a role
that 6LoWPAN NR can perfectly play

>     3) nodes that are sleepy, are not going to be useful as members of
>	an RPL DAG, so really, it's about RPL between the Edge Routers.

In my mind the edge router is usually also the root for a RPL DAG.

I agree that a node that exposes itself as a router must be able to get
packets from nodes, and hold packets for next hops till they are awake
to get them. So obviously, being a router will cause additional drain. 

From there, multiple strategies are possible, and I like the one whereby
the router role can roll - ;) The current draft does not provide means
to automate that strategy. It is also possible to deploy power-savvy
routers to provide a mesh coverage.

The protocol provides the means but how they are used is up to
deployment...

>It seems to me therefore that the only nodes in a 6lowpan that can
>participate in a RPL DAG would be the edge router that runs a
>Whiteboard.

RPL is the prime protocol for a route-over 6LoWPAN network. So we must
make it work, and hopefully it does.

>I think you capture this in your above sentence.   I do not think that
>the 6LoWPAN routers will be speaking NR/NC between themselves along
>their "backbone", but I could be wrong.

Any 6LoWPAN node joins and stays in the LoWPAN using NR/NC with at least
one router. If the node is itself a RPL router, it makes sense that it
joins via its selected parents. Guess what, now we have a perfectly
natural binding for our DAO. Hardly a surprise though seems there are a
number of people that participate to both WGs.

>I also think that some other people may be thinking that many more
>6LoWPAN nodes will participate in RPL than just the edge routers.

Hopefully :)

Pascal