Re: Nearest Common Ancestor Report (XDb1's $1000 Challenge)

On Tue, 18 May 2004 20:14:42 -0300, Leandro Guimaraens Faria Corsetti Dutra wrote:

>Em Tue, 18 May 2004 22:20:46 +0200, Hugo Kornelis escreveu:
>
>> That's probably the reason why I came up with a working solution
>
> It is just that you bothered to. The rest of us gave up on
>Neo the troll a long time ago.

Which would probably have been the wiser thing to do. But when I saw Neo's challenge, which was clearly specifically designed to show off the things XDb1 does best and which uses a standard report, built into XDb1 (and therefor probably heavliy optimized, maybe even down to C or ASM code), I considered it a nice challenge to see if I could get an RDBMS to beat XDb1 at it's own game. It's just that I like a good challenge every now and then.

Unfortunately, this proved out to be no challenge at all. My first, pretty straightforward attempt already managed to beat XDb1.

So far, the exchange with Neo is fun. Neo has already disclosed so many details of XDb1's structure that I now clearly understand *why* XDb1's report whows such horrendous performance.

> And we don't like cross-posting.

You certainly do have a point here.

Best, Hugo

-- 

(Remove _NO_ and _SPAM_ to get my e-mail address)