Tom,
Using something newer than 8.1 would help.
Thank you.
If
CREATE TEMP TABLE ids ( id int ) ON COMMIT DROP;
is created, ids are added to this table and
ids table is used in inner join insted of IN clause or
IN clause is replaced with
... dokumnr IN ( SELECT id FROM ids ) ...
, w
Obviously, most of the total cost (cost=327569, time=39749ms) comes from
two operations in the execution plan:
(a) sequential scan on the 'rid' table (cost=195342, time=26347ms) that
produces almost 3.200.000 rows
(b) hash join of the two subresults (cost=24, time=14000ms)
How many rows are t
"Andrus" <[EMAIL PROTECTED]> writes:
> How to make this query to run fast ?
Using something newer than 8.1 would help.
regards, tom lane
--
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgre
I found that simple IN query on indexed tables takes too much time.
dok and rid have both indexes on int dokumnr columnr and dokumnr is not
null.
PostgreSql can use index on dok or event on rid so it can executed fast.
How to make this query to run fast ?
Andrus.
note: list contain a lot o
