On Wed, Jul 2, 2008 at 9:17 AM, Noel Chiappa <[EMAIL PROTECTED]> wrote: > > routing and addressing are fundamentally inseparable > > True, but addressing is just one aspect of a routing architecture - O() 20% > of the whole, I'd say.
Noel, That rather depends on the architecture. If you take the tack I posted about recently (ephemeral locater-only addresses at layer 3), address assignment, deprecation, revocation and reassignment becomes more like 80% of the routing protocol's job. The rest looks a lot like other routing protocols now but with a fraction of the routes due to very effective address aggregation. On Wed, Jul 2, 2008 at 11:59 AM, Noel Chiappa <[EMAIL PROTECTED]> wrote: > Addresses (routing-names) are the data that path-selection (routing) works > on. As such, they are inherently inseparable. Layer-3 addresses presently describe two characteristics of the endpoint: its network location and its identity. In a clean slate environment, it is not obvious to me that path-selection need know anything about the identity part; it need only know about one of the network locations. If you take identity out of layer 3 entirely and insist that layer 4 use the layer-3 IP address no more than layer 3 uses the layer-2 MAC address, a lot of really interesting things become possible in the routing/addressing space. It occurs to me as we talk about splitting locaters from identity, we may be saying that OSI layer 3 is conceptually wrong. We need to establish a node identity somewhere beneath the transport layer but above the routing at the network layer. Is there a layer missing between 3 and 4? Regards, Bill Herrin -- William D. Herrin ................ [EMAIL PROTECTED] [EMAIL PROTECTED] 3005 Crane Dr. ...................... Web: <http://bill.herrin.us/> Falls Church, VA 22042-3004 -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
