Re: Now I am back, next thing. Final PGS tuning.

From: Greg Smith <gsmith(at)gregsmith(dot)com>
To: Jennifer Trey <jennifer(dot)trey(at)gmail(dot)com>
Cc: "Massa, Harald Armin" <chef(at)ghum(dot)de>, Bill Moran <wmoran(at)potentialtech(dot)com>, pgsql-general(at)postgresql(dot)org
Subject: Re: Now I am back, next thing. Final PGS tuning.
Date: 2009-04-08 18:16:19
Message-ID: alpine.GSO.2.01.0904081316370.22528@westnet.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, 8 Apr 2009, Jennifer Trey wrote:

> shared_buffer = 1024MB # Kept it

As mentioned a couple of times here, this is a really large setting for
Windows. Something like 256MB would work better, and you might even find
some people making a case for 64MB or less on Windows. I don't really
know for sure myself.

> Is the effective cache only the one for the OS ? not for them combined ?

It is sizing the combination of the shared_buffers *plus* what you expect
in the OS buffer cache. I normally look at the size of the OS buffer
cache before the PostgreSQL server is started as a rough estimate here.

> Since I use Java, prepared statements are quite natural.

Prepared statements are not prepared transactions. It's unlikely you've
got any code that uses PREPARE TRANSACTION, so you shouldn't need to
increase max_prepared_transactions.

All three of the above are not really clear in the tuning guide on the
wiki, I'll do an update to improve those sections when I get a minute.

--
* Greg Smith gsmith(at)gregsmith(dot)com http://www.gregsmith.com Baltimore, MD

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Robert Treat 2009-04-08 19:13:27 Re: Are there performance advantages in storing bulky field in separate table?
Previous Message Scott Marlowe 2009-04-08 17:53:55 Re: Now I am back, next thing. Final PGS tuning.