Jim Sander wrote:
> <snip>
>
> For example, you resolve 207.239.68.10 to "trelane.addy.com". You cache
> this name, and when you get a request from 207.239.68.9 you determine
> that 207.239.68.* is "addy.com" and have done with it.
>
> Of course this system isn't perfect since not all domains are broken up
> along class C network boundaries, but it would speed things up somewhat
> (DNS lookups longer than everything else analog does combined) and it's
> close enough since I'd bet that most of those "oddball" sites don't
> originate traffic anyway.
>
> Anyone else think about this and find something I haven't?
>
Actually this probably doesn't work well. First, most non-government U.S.,
unfortunately, sites have one fewer "dot worth" of info than the rest of the
world. Second, many ISPs are still small company originating less the one
class "C" ip set or less than an even number of class "C"s. On top of that as
IP #'s are get used up (or get reserved) there is going to be more and more
trading and aquiring of less than complete class "C" sets. (Until IPv6 is
implemented, companies wanting to aquire more IP's may be left to aquire a set
of 32 or 64 from some company that owns a class "C" but won't be using it.
After all there's only 1,000,000,000 class "C"s available :).
Aside for STEPHEN: Again, how about a page for links to these things? "Analog
Configuration Resources"
--
Jeremy Wadsack
OutQuest Magazine
a Wadsack-Allen publication
--------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
--------------------------------------------------------------------
