[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RAM] Number of DFZ routers - radical improvement of BGP unlikely



I am pretty in line  with Iljitsch. I also think that it is feasonable/reasonable to have the new design in the middle. At least for an introduction period.
When I would talk about 1-2 K FIB entries, I mean of course 1-2 K links,mainly loose links, the more remote, the looser, however precisest in length. I would also call the database "map-based". But because this is also a routing table, the alternative should rather be called "prefix-based".
 
Same remarks about the current stage.
We have had a very long discussion about the dual nature id/loc. We never had a discussion about strict/loose links. I mean  algorithmically determined loose links.
We have adopted the term "stretch" but none of the current models (LISP,lvip,..) has anything to do with it. It takes  map-based models, as to rate them by stretch values, doesn't it ? (I would claim stretch=1 for my map-based concept of course :-)
 
Another fundamental point of discussion is:
Shall a) the IP-header  serve the model or b) shall the model serve the IP-header ?
We have seen that a) is possible (TOS converted to DSCP). Why not again here !?
 
In summary, it is too early to start with the finish.
 
Heiner 
 
In einer eMail vom 07.09.2007 13:27:10 Westeuropäische Normalzeit schreibt iljitsch at muada.com:
On 7-sep-2007, at 3:47, Noel Chiappa wrote:

>> From: Iljitsch van Beijnum <iljitsch at muada.com>
>
>> The question is whether it would be easier to replace it. So far, I
>> can't think of anyone else other than myself speaking out in favor of
>> that, or even entertaining the question seriously. :-)

> Really? :-)

Dangers of caching: information from long ago may disappear unless 
it's refreshed periodically.  (-:

>> Obviously a new protocol would have to be able to interact with 
>> BGP and
>> be not much worse at talking BGP than a native speaker.

> Well, "interact with BGP" and "talk BGP" sound like two very 
> different things
> to me - unless by "talk BGP" you mean "emulate a BGP speaker".

Yes, they are different things: there is the protocol, which would be 
more or less independent from BGP, and the implementation, which 
would almost certainly need to be able to talk both the new protocol 
and talk to BGP speakers. Whether this counts as emulating... don't 
know.

> The thing is that if you have a system with a fundamentally 
> different model
> of the world (e.g. map-based, instead of route-table based), the 
> interface
> between it and BGP is necessarily going to be something of a 
> kludge, and any
> emulation is perforce going to be be something less than stellar.

At the very least a new protocol would have to support the situation 
where the new protocol is spoken in the middle and BGP both on the 
left and on the right. Obviously the BGP speakers may at that point 
also interact through regular BGP, e.g.:

BGP A -- New -- BGP B
       \       /
         BGP C

(Where "new" may be one or more ASes running the new protocol.) So 
the new protocol must be able to emulate BGP processing so well that 
it allows A and B to make (almost) identical decisions as in the case 
where there was no new protocol. For instance, AS paths must be at 
least retained and probably be added to as information travels 
through the new protocol. It gets even more interesting when two 
clouds speaking the new protocol are separated by a BGP cloud.

I don't think an internet-wide map based protocol is feasible, unless 
aggregation is so draconian that the protocol itself becomes largely 
irrelevant and operators will mostly be dealing with traffic 
engineering exceptions rather than the protocol itself. A system 
where there is a map of the "center" of the network (as seen from a 
given vantage point) where mapping to/from prefix tables happens at 
the edges of the map would be very interesting.

_______________________________________________
RAM mailing list
RAM at iab.org
https://www1.ietf.org/mailman/listinfo/ram
 
_______________________________________________
RAM mailing list
RAM at iab.org
https://www1.ietf.org/mailman/listinfo/ram