RE: distributed LDAP Re: performance

["Bengt Hammarlin" <Bengt.Hammarlin@mobeon.com>]

> I  am looking at ways to extend the performance of LDAP on large 
> databases without throwing hardware at the problem. My instinctive 
> solution would be some sort of distributed database.

How is that different from "throwing hardware at the problem?" Obviously

  to have a distributed database you must use even more hardware.

A larger box costs more per MIPS than a smaller and it costs more to
upgrade in small increments.
To increment with a distributed solution you add comodity hardware. You
are also in theory less affected by hardware failures.