Re: radix tree implementation for quota ?

[Sad Clouds <cryintothebluesky%googlemail.com@localhost>]

On Sun, 28 Nov 2010 18:27:07 +0100
Manuel Bouyer <bouyer%antioche.eu.org@localhost> wrote:

> > Wouldn't a hash table work?
> 
> I think it's too dependant on uid distribution, or even how much uid you
> have. a tree scales and adapts better.

I'm not sure I follow, if you size the hash table correctly and pick a good 
hash function, it will give you very good results.

You say that current implementation uses an simple array, indexed by uid. This 
would be ideal for a hash table. How would a tree scale better?