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

Re: [RRG] Taxonomy: 25 questions - Heiner Hummel's proposal



Hi Robin,
 
By ingress node and egress node I mean ingress router and egress router.
Let's assume the ingress router has added the geographical coordinates as optional parameter.
Those routers which do not understand my architecture proceed just like now (quasi watch for the direction signs aside of the road), while those routers which know my concept  do the forwarding based on the geographical coordinates if provided, or just like today if not provided (quasi  either watch for the direction signs, or listen to the navigating voice).
 
There is one point left open:How can my concept evolve while there are a lot of routers in-between which do not contribute so that the routers can aquire their specific hierarchichally sparsed network view ? But for this sub-problem there are patterns, e.g. MPLS/VPN RFC2547, how knowlegeable routers may learn about each other, although they are not immediate neighbors. Yes, needs further study, but precisely here, all who know better than me the currently deployed protocols, can help for sure.
 
BTW (incentive): You may look at the graph on my web-site and see the shown multipath capability (which can even be extended) for a strict-links based intra-domain network. The same capability could be provided for the combined intra/inter-domain internet with strict intra-domain links, and with loose inter-domain links, which are the looser and concurrently the stablier the more remote they are.
 
 
Heiner