Re: [Roll] updating DAO caches (was Re: Something to ADD)

["Julien Abeille (jabeille)" <jabeille at cisco.com>]

Hi Richard,

Pure source routing is supported in the draft. I think it's much less
random to maintain this with specific messages (non stored DAOs) than
with application traffic. This way the root always has a route to every
node. What are you missing for your scenario?

Best,
Julien

> -----Original Message-----
> From: roll-bounces at ietf.org [mailto:roll-bounces at ietf.org] On 
> Behalf Of Richard Kelsey
> Sent: jeudi 19 novembre 2009 16:50
> To: JP Vasseur (jvasseur)
> Cc: roll at ietf.org
> Subject: Re: [Roll] updating DAO caches (was Re: Something to ADD)
> 
> > From: JP Vasseur <jvasseur at cisco.com>
> > Date: Thu, 19 Nov 2009 15:49:15 +0100
> > 
> > 
> > On Nov 19, 2009, at 3:41 PM, Richard Kelsey wrote:
> > 
> > > The issue isn't whether or not we need good P2P routing.
> > > The issue is whether or not the DAO mechanism does a good 
> enough job 
> > > to be worth the effort.
> > >
> > > RPL ignores P2P when choosing parents.  Getting a good 
> P2P route out 
> > > of RPL is a matter of luck, not design.
> > 
> > First, you can certainly tune the OF to increase the P2P quality.
> 
> I haven't thought about it a lot, but it isn't at all clear 
> to me how you would do this.
> 
> I believe that we can get good P2P routes using RPL as a 
> base.  Doing so will require more than just coming up with an 
> appropriate OF.
> 
> > Second, in the VERY worst non probable case you would 
> transit via the 
> > root.
> 
> We clearly have very different images of the DAGs that RPL 
> will produce.  The DAG formation mechanism works to maximize 
> downward fanout, which in turn maximizes the number of P2P 
> routes that go through the root even if all nodes store DAOs. 
>  For example, if the root has N children with roughly equal 
> subgraphs, then each node can reach only 1/N of the network 
> without through the root.
> 
> > Last but not least, without DAO how do you send traffic 
> outward?  Need 
> > to wait until you receive a packet, then record route and do source 
> > routing?
> 
> My point is that your argument in favor of requiring multiple 
> DAO caches is also an argument that DAO is not adequate for 
> P2P routing.
> 
> You are claiming that we have to cache DAOs in non-root nodes 
> because otherwise there is too much traffic through the root. 
>  If there is too much traffic through the root without the 
> caching, then there is almost certainly too much even with 
> the caching.  Having nodes cache DAOs cannot be counted on to 
> greatly reduce the traffic throught the root.
> 
> If the DAO mechanism meets the requirements for P2P, then it 
> meets those requirements with DAOs only cached at the root.
> 
>                             -Richard Kelsey 
> _______________________________________________
> Roll mailing list
> Roll at ietf.org
> https://www.ietf.org/mailman/listinfo/roll
>