From: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl> |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, listuser(at)peternixon(dot)net, pgsql-general(at)postgresql(dot)org |
Subject: | Re: Delete large amount of records and INSERT (with indexes) |
Date: | 2003-04-15 14:47:03 |
Message-ID: | 200304151447.h3FEl3p09572@candle.pha.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Alvaro Herrera wrote:
> On Wed, Apr 09, 2003 at 12:59:01PM -0400, Tom Lane wrote:
> > Peter Nixon <listuser(at)peternixon(dot)net> writes:
> >
> > > I then reran my import script and found that I was getting approximately 1
> > > INSERT every 30 secconds!!! although SELECTS were working relatively
> > > quickly.
> >
> > I think your mistake was to do VACUUM ANALYZE while the table was empty.
> > That led the planner to generate plans suitable for a very small table
> > --- seqscans instead of indexscans, for example.
>
> But he _is_ getting good performance for SELECT, so the plans are not
> that bad. Only insertion is slow. Can this be related to the btree
> index bloating issue you fixed on january?
It is my understanding that we only allowed indexes to reuse space after
VACUUM, not shrink the index size during VACUUM FULL, so the index could
still be huge.
--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073
From | Date | Subject | |
---|---|---|---|
Next Message | Oleg Bartunov | 2003-04-15 14:48:40 | Re: Appending to an array |
Previous Message | Johnson, Shaunn | 2003-04-15 14:41:44 | import fixed width files into PostgreSQL |