| From: | "Sergey Konoplev" <gray(dot)ru(at)gmail(dot)com> |
|---|---|
| To: | "pgsql-general(at)postgresql(dot)org >> PG-General Mailing List" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Explain's estimation differs from real count enormously |
| Date: | 2008-10-23 12:29:21 |
| Message-ID: | c3a7de1f0810230529s3cdbaa21ha293013dbdf89dcb@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
>
> EXPLAIN ANALYZE of the query shows strange estimated rows count:
>
> EXPLAIN ANALYZE SELECT * FROM objonmap
> WHERE 1=1
> AND box(om_point, om_point) <@ box(point(-55.97398205077, -100),
> point(82.166446008477, 180))
> AND (om_flag OR om_ref_id IN (69000001071, 69000001513,
> 69000001092, 69000001091))
> AND om_state = 1;
>
One more thing - removing condition "AND box(om_point, om_point) <@
box(point(-55.97398205077, -100), point(82.166446008477, 180))" leads
to good (right) estimation.
BTW, pg 8.3.3 behave the same.
--
Regards,
Sergey Konoplev
--
PostgreSQL articles in english & russian
http://gray-hemp.blogspot.com/search/label/postgresql/
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Thomas Guettler | 2008-10-23 12:44:20 | Re: Shopping cart |
| Previous Message | Mikkel Hgh | 2008-10-23 12:20:49 | Re: Database design for separate tsearch table |