Re: [Roll] AMI applicability: storing/non-storing.

Hi Don,

Sorry for the delayed reply...was out of commission for awhile.

Yes, I'm aware of the existing spec'd forms of storing and non-storing, however,  the "hybrid storing" mode is not hybrid with regards to unicast, but just simply listening to multicast joins from your children and storing those group addresses.   I don't think this is that complex -- it would benefit the non-flooding method that I referred to in the research paper, without adding too much RAM burden for the routing motes in the network.  If my assumptions regarding the number of multicast groups that might be active in an LLN is not correct, then that would have some application (may be appropriate for some networks, but not others) affinity.  

Randy

-----Original Message-----
From: roll-bounces@ietf.org [mailto:roll-bounces@ietf.org] On Behalf Of Don Sturek
Sent: Tuesday, November 05, 2013 2:16 PM
To: Routing Over Low power and Lossy networks
Subject: Re: [Roll] AMI applicability: storing/non-storing.

Hi Randy,

"Hybrid storing" modes were looked into awhile ago in ROLL and abandoned due to complexity.  Currently, only storing and non-storing is supported in a specific DODAG instance.

I think the topic of multicast support is a bit separate.   I think you
forwarded me a research paper that used storing mode to analyze multicast
group member reachability.   That is an interesting topic when considering
multicast forwarding approaches that don't use flooding (although, I would expect that storing mode would be used with the approach since you need all of that routing information, plus multicast group membership, to make forwarding decisions)

Don

On 11/5/13 10:59 AM, "Turner, Randy" <Randy.Turner@landisgyr.com> wrote:

>I agree that non-storing mode addresses the primary AMI use-case where 
>endpoints push data up and through the DODAG root.
>
>Could there be a "hybrid-storing" mode that doesn't use as much RAM to 
>contain routes to all nodes in a sub-tree? Instead of traditional 
>storing mode, could a node store just multicast memberships in all nodes "below"
>a particular mote ?  Assuming there would much fewer multicast 
>memberships than would be required for RFC 6550 storing mode ?  It's a 
>variant of storing+multicast, without the traditional storing mode RAM 
>usage.
>
>I'm still looking at the MPL document, so apologies if that's already 
>discussed.
>
>Randy
>
>-----Original Message-----
>From: roll-bounces@ietf.org [mailto:roll-bounces@ietf.org] On Behalf Of 
>Michael Richardson
>Sent: Tuesday, November 05, 2013 1:50 PM
>To: Routing Over Low power and Lossy networks
>Subject: Re: [Roll] AMI applicability: storing/non-storing.
>
>
>Popa, Daniel <Daniel.Popa@itron.com> wrote:
>    Daniel> However, my point is that these challenges are mostly 
>related to
>    Daniel> RPL storing mode of operation. As a consequence, I am OK to
>    Daniel> narrow down the scope of the applicability statement for 
>AMI to
>    Daniel> only RPL non-storing mode of operation, as I stated in my
>    Daniel> previous email. As for having different specs for different 
>use
>    Daniel> cases, I expect that by focusing only on RPL non-storing 
>mode a
>    Daniel> single AMI applicability statement will be able to cover
>    Daniel> efficiently "all" the reasonable requirements, for "all" use
>    Daniel> cases.
>
>Ah, I did not understand that from your previous email.
>Yes, please.  Pick non-storing, and lets' proceed!!!
>
>--
>]               Never tell me the odds!                 | ipv6 mesh
>networks [
>]   Michael Richardson, Sandelman Software Works        | network
>architect  [
>]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails
>   [
>
>
>P PLEASE CONSIDER OUR ENVIRONMENT BEFORE PRINTING THIS EMAIL.
>
>This e-mail (including any attachments) is confidential and may be 
>legally privileged. If you are not an intended recipient or an 
>authorized representative of an intended recipient, you are prohibited 
>from using, copying or distributing the information in this e-mail or 
>its attachments. If you have received this e-mail in error, please 
>notify the sender immediately by return e-mail and delete all copies of 
>this message and any attachments. Thank you.
>_______________________________________________
>Roll mailing list
>Roll@ietf.org
>https://www.ietf.org/mailman/listinfo/roll

_______________________________________________
Roll mailing list
Roll@ietf.org
https://www.ietf.org/mailman/listinfo/roll