Re: Performance - Mailing list pgsql-general

Here are more info. Sorry, I thought taht because this case is very simple,
it was not necesary, but ....


50000 ROWS  aprox

QUERY:
SELECT * FROM d_cue WHERE d_cue.clave = '$cue'

INDEX:
create index d_cue_clave on d_cue (clave);

EXPLAIN:
ra1999=> explain select * from d_cue where clave='9400001';
NOTICE:  QUERY PLAN:

Seq Scan on d_cue  (cost=0.00..3738.62 rows=1 width=544)


TABLE STRUCTURE:
CREATE TABLE "d_cue" (
 "clave" character(7),
 "nombre" character varying(60),
 "calle" character varying(45),
 "referencia" character varying(60),
 "telefono" character varying(15),
 "cp" int4,
 "zona" int4,
 "fraccion" character(2),
 "radio" character(2),
 "cooperadora" int4,
 "d_cooperadora" character varying(50),
 "confesional" int4,
 "d_confesional" character varying(50),
 "arancel" int4,
 "d_arancel" character varying(50),
 "categoria" int4,
 "d_categoria" character varying(50),
 "permanencia" int4,
 "d_permanencia" character varying(50),
 "alternancia" int4,
 "d_alternancia" character varying(50),
 "cod_jur" character varying(15),
 "matric" character,
 "periodo_func" int4,
 "d_periodo_func" character varying(50),
 "var1" character varying(15),
 "var2" character varying(15),
 "var3" character varying(15),
 "var4" character varying(15),
 "var5" character varying(15),
 "var6" character varying(15),
 "vat1" character(6),
 "d_vat1" character varying(50),
 "vat2" character(6),
 "d_vat2" character varying(50),
 "vat3" character(6),
 "d_vat3" character varying(50),
 "vat4" character(6),
 "d_vat4" character varying(50),
 "vat5" character(6),
 "d_vat5" character varying(50),
 "vat6" character(6),
 "d_vat6" character varying(50),
 "direle" character varying(30),
 "barrio" character varying(40),
 "nuevo" character,
 "d_nuevo" character varying(50),
 "turnos" character varying(255),
 "tipos" character varying(255),
 "ciclos" character varying(255),
 "f_nf" character(10),
 "agregado" int4);

Diego Schvartzman
Email: diego.schvartzman@usa.net
ICQ# 1779434
----- Original Message -----
From: Alfred Perlstein <bright@wintelcom.net>
To: Dustin Sallings <dustin@spy.net>
Cc: Charles Tassell <ctassell@isn.net>; Diego Schvartzman
<dschvar@yahoo.com>; Lista PGSQL <pgsql-general@postgresql.org>
Sent: Tuesday, May 16, 2000 1:14 PM
Subject: Re: [GENERAL] Performance


> > # At 04:56 PM 5/15/00, Diego Schvartzman wrote:
> > # >I have an application via PHP. For example, a SELECT query that must
return
> > # >one and only one row, with a where clause with and index (I droped it
and
> > # >created again) that took about 3 seconds (v6.5.3), now (v7.0.0) takes
about
> > # >15 seconds. Really I don't lnow what is happening. Same hardware,
same php
> > # >version, etc etc.
> > # >
>
> * Dustin Sallings <dustin@spy.net> [000516 02:26] wrote:
> > On Mon, 15 May 2000, Charles Tassell wrote:
> >
> > I ran into this exact problem, and it was *very* significant on a
> > 15M row table I have.  :)  It didn't seem to want to use the index, even
> > freshly created, without a vacuum analyze.
>
> grrrr....
>
> FOR THE LAST TIME, THESE BUG REPORTS ARE PRETTY MUCH **USELESS**
> TO THE DEVELOPERS UNLESS YOU GIVE:
>
> THE TABLE STRUCTURE,
> THE QUERY, AND
> THE OUTPUT OF 'EXPLAIN'
>
> Just because someone is a database guru doesn't mean they are also
> clairvoyant. :)
>
> thanks,
> -Alfred
>