Re: [Roll] New proposed charter

[Randy Turner <[e-mail]>]

Hi Michael,

RFC 6550 mentions the possibility of hybrid storing/non-storing approaches, but only mentions the possibility and declares it out of scope.  I do think it would be nice to re-visit this topic and see if there’s something there of value.

The additional type of routing I was referring to was similar to this “hybrid”, but specific to multicast. Something like “Non-storing with multicast-storing”

This would be basically a non-storing mode of operation, except a node would  just store multicast memberships that have been declared - if someone joins a multicast group, you store this…since memberships can be aggregated, it shouldn’t mean too much of a burden on the amount of memory required to store multicast routes.


Randy


> On Dec 11, 2015, at 4:23 PM, Michael Richardson <[e-mail]> wrote:
> 
> 
> Randy Turner <[e-mail]> wrote:
>> We might want to consider rolling items from 6553, 6551, etc. back into
>> an overall “RPL Protocol” RFC – not everything, but items that we have
>> found in implementation to be a requirement for a workable RPL
>> implementation.
> 
> That's entirely doable if we have the editorial cycles, and it's included in:
> 
>    ines>     Over time, additional issues have come up, including issues in the
>    ines> data plane of when to use IPinIP headers, and how to compress them, as
>    ines> well as the standardization of a mixed storing/non-storing mechanism.
> 
> So, we'd be *updating* 6554,6553,6551, etc. in explaining how to compress
> things.
> 
>    ines> aim of eventually revising and advancing RFC
>    ines> 6550 to Internet Standard.
> 
> And this item includes the possibility that we could combine multiple
> documents as we go to IS.
> When it says "RFC6550" it should mean the entire RPL document set.
> So perhaps that part should be more explicit?
> 
>> Also, I would like to consider other options for routing besides
>> strictly storing and non-storing, or mixed storing/non-storing
> 
> Explicitely already mentioned: we didn't pick a solution in the charter
> above:      **standardization of a mixed storing/non-storing mechanism**
> 
>> Randy
> 
> So, are you in favour of this change, or can you suggest specific text?
> 
> --
> Michael Richardson <[e-mail]>, Sandelman Software Works
> IETF ROLL WG co-chair.    http://datatracker.ietf.org/wg/roll/charter/
>