Re: ason-routing-reqts: issue 2 architecture

[Dimitri.Papadimitriou@alcatel.be]

don,

> So Adrian took us back to requirements. But I think we agree there are at
> least 2 ways to do it and both may be required in the end.

(1) that's fine, we reached the first discussion point which is how 
local representation of such construct can be done and achieved, it 
clearly shows there are several ways (btw, i assume here that we agreed 
on the "external" link representation)

(2) the second point of discussion is how much of this information is 
really needed externally (ie the common set of information) and a common 
semantic to what it represents to make the control plane working in a 
standardized and interoperable way (note: point (1) discussion showed 
that what's internal will be more elaborated than what is externally 
known and this in any case)

-> it seems that the following responds to the question: "A collection 
of links and nodes such as a subnetwork or RA must be able to represent 
itself to the wider network as a single logical entity with only its 
external links visible to the topology database.)" can we agree on this one

(3) and then, but i don't think we will be capable to enter into that 
discussion for the time being since back to the functional requirements:
- (3a) what's needed to represent this information in the routing
  protocol
- (3b) and what's needed as routing mechanism to exchange this
  information

thanks,
- dimitri.
> Don 
>
> <snip>

--
Papadimitriou Dimitri
E-mail : dimitri.papadimitriou@alcatel.be
E-mail : dpapadimitriou@psg.com
Webpage: http://psg.com/~dpapadimitriou/
Address: Fr. Wellesplein 1, B-2018 Antwerpen, Belgium
Phone  : +32 3 240-8491