Subject: Re: Performance
Mime-Version: 1.0 (Apple Message framework v1082)
Content-Type: text/plain; charset=us-ascii
From: Ogden <lists@darkstatic.com>
In-Reply-To: <20110412171855.GA14292@tux>
Date: Tue, 12 Apr 2011 12:23:15 -0500
Cc: pgsql-performance@postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <FC3A3A2B-3ECB-41BA-8F94-356D6FED3695@darkstatic.com>
References: <DF2D5436-117C-4D02-9B3C-A55723B7DDE1@darkstatic.com>
	<20110412171855.GA14292@tux>
To: Andreas Kretschmer <akretschmer@spamfence.net>


On Apr 12, 2011, at 12:18 PM, Andreas Kretschmer wrote:

> Ogden <lists@darkstatic.com> wrote:
>=20
>> I have been wrestling with the configuration of the dedicated Postges =
9.0.3
>> server at work and granted, there's more activity on the production =
server, but
>> the same queries take twice as long on the beefier server than my mac =
at home.
>> I have pasted what I have changed in postgresql.conf - I am wondering =
if
>> there's any way one can help me change things around to be more =
efficient.
>>=20
>> Dedicated PostgreSQL 9.0.3 Server with 16GB Ram
>>=20
>> Heavy write and read (for reporting and calculations) server.=20
>>=20
>> max_connections =3D 350=20
>> shared_buffers =3D 4096MB =20
>> work_mem =3D 32MB
>> maintenance_work_mem =3D 512MB
>=20
> That's okay.
>=20
>=20
>>=20
>>=20
>> seq_page_cost =3D 0.02                    # measured on an arbitrary =
scale
>> random_page_cost =3D 0.03=20
>=20
> Do you have super, Super, SUPER fast disks? I think, this =
(seq_page_cost
> and random_page_cost) are completly wrong.
>=20

No, I don't have super fast disks. Just the 15K SCSI over RAID. I find =
by raising them to:

seq_page_cost =3D 1.0
random_page_cost =3D 3.0
cpu_tuple_cost =3D 0.3
#cpu_index_tuple_cost =3D 0.005           # same scale as above - 0.005
#cpu_operator_cost =3D 0.0025             # same scale as above
effective_cache_size =3D 8192MB=20

That this is better, some queries run much faster. Is this better?

I will find the archive and post.=20

Thank you

Ogden