>-----Original Message----- >From: Scott Brim [mailto:[EMAIL PROTECTED] >Sent: Sunday, November 30, 2008 9:56 PM >To: Templin, Fred L >Cc: RRG >Subject: Re: [rrg] The Internet DFZ as a giant NBMA "link" > >Excerpts from Templin, Fred L on Sun, Nov 30, 2008 01:49:12PM -0800: >> I have not seen this widely discussed, but in the map/encaps >> approaches the Internet DFZ is essentially viewed as a giant NBMA >> "link" that connects all *TRs. In other words, all *TRs on the >> virtual link can reach each other in a single IP hop and can use >> neighbor discovery mechanisms such as IPv6 ND. This also paves the >> way for new ND mechanisms such as SEND. > >How would this scale? We're considering O(10^9) prefixes and rather >more xTRs.
Simply add more xTRs and make sure their RLOCs are published in the mapping database. Thanks - Fred [EMAIL PROTECTED] >> This model exactly parallels that which was established in ISATAP >> [RFC5214], and has been further developed in RANGER/VET/SEAL. I made >> an effort to articulate this back in 2002 in a note titled: "The >> Inside-Out ISATAP Model": >> >> http://www.ops.ietf.org/lists/v6ops/v6ops.2002/msg00657.html >> >> but we seem to now be coming back around to the same concept. The >> main point of this note therefore is to (re)introduce the useful >> concept of the Internet DFZ as a giant NBMA "link" that connects >> routers the same as for any link. > >I'll take a look again. > >Scott _______________________________________________ rrg mailing list [email protected] https://www.irtf.org/mailman/listinfo/rrg
