Re: Sustained inserts per sec ... ?

From: Mike Rylander <mrylander(at)gmail(dot)com>
To: Simon Riggs <simon(at)2ndquadrant(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Sustained inserts per sec ... ?
Date: 2005-04-04 23:28:25
Message-ID: b918cf3d05040416287bf95da2@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

If I'm getting the point of this thread correctly, have a huge amount
of data in one table degrades INSERT/COPY performance even with just a
PKEY index. If that's about the size of it, read on. If not, ignore
me because I missed something.

On Apr 4, 2005 10:44 PM, Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
> Before I start to tunnel-vision on a particular coincidence...
>

Don't worry too much about tunnel vision. I see the same thing every
day with multi-million row tables. The bigger the table gets (with
only a pkey index) the slower the inserts go. If I start over
(truncate, drop/create table), or if I point the initial load at a new
table, everything gets speedy. I've always figured it was a function
of table size and learned to live with it...

> How much memory have you got on the system?

On mine, 16G

> How much of that have you allocated to various tasks?

shared buffers: 15000

> What else is happening on your system?

Nothing on mine.

> Tell us more about disk set-up and other hardware related things.

6-disk RAID10 on a Compaq SmartArray 6404 with 256M BB cache, WAL on
2-disk mirror on built in SmartArray5 controller.

--
Mike Rylander
mrylander(at)gmail(dot)com
GPLS -- PINES Development
Database Developer
http://open-ils.org

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Steven Rosenstein 2005-04-04 23:40:30 Bulk COPY end of copy delimiter
Previous Message Simon Riggs 2005-04-04 22:44:00 Re: Sustained inserts per sec ... ?