Re: 7.3 vs 7.2 - different query plan, bad performance

On Sun, Jun 15, 2003 at 09:48:08PM +0200, Manfred Koizar wrote:
> On Fri, 13 Jun 2003 20:45:06 +0200, Ryszard Lach <siaco@autograf.pl>
> wrote:
> >I have a problem with performance after upgrading from 7.2 to 7.3.
>
> Try
>     VACUUM ANALYSE;
>
> and then re-run your query.  If it is still slow, post the new EXPLAIN
> ANALYSE output here.
>

Hm. I've tried it too. I don't see a big difference:

siaco=# explain analyze  select count(*) from v_c;
                                                        QUERY PLAN

---------------------------------------------------------------------------------------------------------------------------
 Aggregate  (cost=210.83..210.83 rows=1 width=8) (actual time=5418.09..5418.09 rows=1 loops=1)
   ->  Subquery Scan v_c  (cost=28.40..197.63 rows=5281 width=8) (actual time=4.59..5414.13 rows=5281 loops=1)
         ->  Hash Join  (cost=28.40..197.63 rows=5281 width=8) (actual time=4.58..5407.73 rows=5281 loops=1)
               Hash Cond: ("outer".id = "inner".id)
               ->  Seq Scan on b t1  (cost=0.00..76.81 rows=5281 width=4) (actual time=0.01..9.68 rows=5281 loops=1)
               ->  Hash  (cost=24.32..24.32 rows=1632 width=4) (actual time=3.29..3.29 rows=0 loops=1)
                     ->  Seq Scan on a t2  (cost=0.00..24.32 rows=1632 width=4) (actual time=0.01..1.88 rows=1632
loops=1)
               SubPlan
                 ->  Aggregate  (cost=28.41..28.41 rows=1 width=0) (actual time=1.02..1.02 rows=1 loops=5281)
                       ->  Seq Scan on a t3  (cost=0.00..28.40 rows=3 width=0) (actual time=0.76..1.01 rows=1
loops=5281)
                             Filter: (parent_id = $0)
 Total runtime: 5433.65 msec


--
"First they ignore you. Then they laugh at you. Then they
fight you. Then you win." - Mohandas Gandhi.