| From: | David Helgason <david(at)uti(dot)is> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Mike Mascari <mascarm(at)mascari(dot)com>, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: I got bit by that darn GEQO setting again... |
| Date: | 2004-01-17 06:32:08 |
| Message-ID: | E01AEE43-48B6-11D8-BAA4-000A9566DA8A@uti.is |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
> Mike Mascari <mascarm(at)mascari(dot)com> writes:
>> But just as a quick notice to those upgrading from 7.3 to 7.4 with
>> fully
>> normalized databases requiring > 11 joins, the GEQO setting can be a
>> killer...
>
> Uh ... dare I ask whether you think it's too high? Or too low?
Just a data point: With a fresh 7.4 and
geqo=on
geqo_threshold=11
geqo_generations=0
geqo_effort=1
geqo_pool_size=0
geqo_selection_bias=2
a 12 table join was taking a whole second to plan until I manually
connected the tables (now it's in the order of a few ms).
I figure geqo had kicked in at that point.
d.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Mike Mascari | 2004-01-17 07:48:51 | Re: I got bit by that darn GEQO setting again... |
| Previous Message | Tom Lane | 2004-01-17 06:13:42 | Re: I got bit by that darn GEQO setting again... |