In response to "Glen Edmonds" <Glen.Edmonds at tafmo.com>:

> > You're arguments are very interesting.  When will you be releasing
> your
> code improvements?
> 
> Bill's comment is fair - my note was not constructive.  My apologies.
> 
> I should have added:
> 
> 1. Is there any plan to enhance Slony to become a clustering solution?

First off, you're apparently using the term "cluster" to mean something
very specific.  I point this out because your use of it is _not_ typical
in my experience, and simply saying "clustering" without further
qualification of what you think that means is going to lead to confusion
in many conversations.

To my knowledge, there is no desire or intention to change the design
of Slony I.  It does an excellent job of what it was designed to do,
so _I_ can't see any reason to change it radically, and I don't know
of anyone who does.

> 2. If not, are there any other "official" open source projects for
> clustering solution to postgres?  If not, would anyone be interested in
> starting one with me?

What do you mean by "official"?

pgpool is the typical solution for synchronous multi-master replication,
if that's what you're looking for.

Slony-II attempted to create asynchronous multi-master replication, but
my understand was that the development team ran into serious performance
problems with the design, and the project has been stalled for quite a
while.  Postgres-R has picked up that banner, but I've been a little
out of touch and couldn't tell you how far along they are.

-- 
Bill Moran
Collaborative Fusion Inc.
http://people.collaborativefusion.com/~wmoran/

wmoran at collaborativefusion.com
Phone: 412-422-3463x4023