Re: [HACKERS] Not enough memory for complex join

On Thu, 4 Mar 1999, Hannu Krosing wrote:
> > I played with the query, and found I can remove (AND d.city_id = 2). The
> > query executes pretty fast (my database is small). I tried to get
> > d.city_id:
> >
> > SELECT p.subsec_id, d.city_id
> > FROM positions p, central cn, shops sh, districts d
> > WHERE cn.pos_id = p.pos_id
> > AND cn.shop_id = sh.shop_id AND sh.distr_id = d.distr_id
> > AND cn.date_i >= current_date - '7 days'::timespan
> >
> > but postgres eats all memory and dies :(((
>
> What does EXPLAIN say ?

EXPLAIN SELECT p.subsec_id, d.city_id
FROM positions p, central cn, shops sh, districts d
WHERE cn.pos_id = p.pos_id
AND cn.shop_id = sh.shop_id AND sh.distr_id = d.distr_id
AND cn.date_i >= current_date - '7 days'::timespan
;
NOTICE: QUERY PLAN:

Nested Loop (cost=0.00 size=1 width=18)
-> Nested Loop (cost=0.00 size=1 width=14)
-> Merge Join (cost=0.00 size=1 width=10)
-> Seq Scan (cost=0.00 size=0 width=0)
-> Sort (cost=0.00 size=0 width=0)
-> Seq Scan on districts d (cost=0.00 size=0 width=4)
-> Seq Scan (cost=0.00 size=0 width=0)
-> Sort (cost=0.00 size=0 width=0)
-> Seq Scan on shops sh (cost=0.00 size=0 width=6)
-> Seq Scan on central cn (cost=0.00 size=0 width=4)
-> Seq Scan on positions p (cost=0.00 size=0 width=4)

EXPLAIN

> --------------------
> Hannu
>

Oleg.
----
Oleg Broytmann http://members.xoom.com/phd2/ phd2(at)earthling(dot)net
Programmers don't die, they just GOSUB without RETURN.