Re: [PERFORMANCE] Stored Procedures

I don't think pgpool is what you need. If I understand pgpool
correctly, pgpool lets you pool multiple postgres servers together. You
are just looking for database connection pooling.

A simple connection pool is basically just an application wide list of
connections. When a client needs a connection, you just request a
connection from the pool. If there is an unused connection in the pool,
it is given to the client and removed from the unused pool. If there is
no unused connection in the pool, then a new connection is opened. When
the client is done with it, the client releases it back into the pool.

You can google for 'database connection pool' and you should find a
bunch of stuff. It's probably a good idea to find one already written.
If you write your own you have to make sure it can deal with things like
dead connections, synchronization, and maximum numbers of open
connections.

Dave

-----Original Message-----
From: pgsql-performance-owner(at)postgresql(dot)org
[mailto:pgsql-performance-owner(at)postgresql(dot)org] On Behalf Of Marcos
Sent: Monday, January 23, 2006 7:27 AM
To: Markus Schaber
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: [PERFORM] [PERFORMANCE] Stored Procedures

Hi Markus

> You really should consider using a connection pool (most web
application
> servers provide pooling facilities) or some other means to keep the
> connection between several http requests.

Yes. I'm finding a connection pool, I found the pgpool but yet don't
understand how it's work I'm go read more about him.

Thanks

Marcos

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org