Re: Send-n, rat holes and real issues
Duane <[email protected]>
| Newsgroups | gmane.ietf.enum |
|---|---|
| Message-ID | <[email protected]> |
[email protected] wrote: > into something at least two orders of magnitude more complicated than a > DNS-based system, for little additional gain. So that's the crux of the argument, it's hard so we'll take the easy way out? No, there most certainly is gain, for starters it can be dealt with in a decentralised manner similar to that of BGP, tier-1 providers allow their customers to announce prefix lengths and filter everything else. Where people should stick to a fixed length to minimise the size of the database OR where there is no fixed length some sort of grouping. People are thinking of this from the dialled digits side only but I'm sure there could be a way to group based on the prefix length as well to minimise the data needing to be stored or broadcast or .... Again this is only about prefix lengths not routing information, and is most certainly not a routing protocol. It has a similar benefit as DNS of being decentralised, but every 'node' would have a cached copy of the information so depending on the equipment and the coding, lookup times would be in the sub-ms, which would reduce post dial delay, where as a few DNS server or network or ...... outages could leave people with a bad experience from extended post dial delays. -- Best regards, Duane