Re: Send-n, rat holes and real issues
Duane <[email protected]>
| Newsgroups | gmane.ietf.enum |
|---|---|
| Message-ID | <[email protected]> |
Edward Lewis wrote: > DNS scales larger than BGP in the sense that the data held by DNS does > not travel like the data held by BGP. Let's say you have 10,000,000 > phone numbers with number portability. But we're not talking about routing, we're talking about prefixes. > I think the point Otmar is making is that when you look at a IPv4 > address, the first N bits determine where all the 32-N bit addresses > are. I.e., a 10.10.10/24 route (if none are more specific) will be the > path for all 256 numbers in 10.10.10.0-255. But for +1-571-555, not all > 10,000 below what will be routed the same way. Exactly, the problem is very similar, you can list the average case, Brian's 99% and then the 1% exception you simply fill in the records as needed. > The obsevation of Otmar that I think is great is that telephone numbers, > when backed by number portability, are no longer network addresses but > are network names. Yet, an IP address is just an address, not a name. > A lookup service (like DNS) or a directory/database deals with names > better than addresses, a routing service deals with addresses better > than names. Number portability has *NOTHING* to do with prefix lengths. unless the owner suddenly decides to increase the number space. -- Best regards, Duane _______________________________________________ enum mailing list [email protected] https://www.ietf.org/mailman/listinfo/enum
signature.asc
(application/pgp-signature, 189 B)
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFIcrZGEpWoZMmo/c8RAmAiAKCWvjWG7DNx6amTw+GzA7LalPw6cwCgpKng vZlT2lNZ2CtHT2SNSjjJsi0= =CQC9 -----END PGP SIGNATURE-----