From: | Robert Creager <Robert_Creager(at)LogicalChaos(dot)org> |
---|---|
To: | Rod Taylor <pg(at)rbt(dot)ca> |
Cc: | Brock Henry <brock(dot)henry(at)gmail(dot)com>, Postgresql Performance <pgsql-performance(at)postgresql(dot)org> |
Subject: | Re: Insert performance, what should I expect? |
Date: | 2004-10-20 16:45:27 |
Message-ID: | 20041020104527.14927c06@thunder.mshome.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
When grilled further on (Tue, 19 Oct 2004 22:12:28 -0400),
Rod Taylor <pg(at)rbt(dot)ca> confessed:
> > I've done some manual benchmarking running my script 'time script.pl'
> > I realise my script uses some of the time, bench marking shows that
> > %50 of the time is spent in dbd:execute.
> >
> 1) Drop DBD::Pg and switch to the Pg driver for Perl instead (non-DBI
> compliant) which has functions similar to putline() that allow COPY to
> be used.
COPY can be used with DBD::Pg, per a script I use:
$dbh->do( "COPY temp_obs_$band ( $col_list ) FROM stdin" );
$dbh->func( join ( "\t", @data ) . "\n", 'putline' );
$dbh->func( "\\.\n", 'putline' );
$dbh->func( 'endcopy' );
With sets of data from 1000 to 8000 records, my COPY performance is consistent
at ~10000 records per second.
Cheers,
Rob
--
10:39:31 up 2 days, 16:25, 2 users, load average: 2.15, 2.77, 3.06
Linux 2.6.5-02 #8 SMP Mon Jul 12 21:34:44 MDT 2004
From | Date | Subject | |
---|---|---|---|
Next Message | Manfred Spraul | 2004-10-20 16:51:49 | Re: futex results with dbt-3 |
Previous Message | Josh Berkus | 2004-10-20 16:38:51 | Re: OS desicion |