Externalizing the limit has improved the speed a lot. Distinct is half a second faster than group by. http://explain.depesz.com/s/vP1 with tmp as ( select distinct tr.nr as tnr , tr.time_end as tend , c.id_board as cb , c.id_board_mini as cbm , ti.id_test_result as itr from test_item ti , test_result tr , component c , recipe_version rv where ti.id_test_result = tr.id and ti.id_component = c.id and tr.id_recipe_version = rv.id and (rv.id_recipe in ('6229bf04-ae38-11e1-a955-0021974df2b2')) and tr.time_end <> cast('1970-01-01 01:00:00.000' as timestamp) and tr.time_begin >= cast('2012-10-27 08:00:17.045' as timestamp) and ti.type = 'Component' --group by tr.nr , tr.time_end , c.id_board , c.id_board_mini , ti.id_test_result order by tr.time_end asc) select * from tmp limit 10000 -- View this message in context: http://postgresql.1045698.n5.nabble.com/Slow-query-where-am-I-going-wrong-tp5730015p5730185.html Sent from the PostgreSQL - performance mailing list archive at Nabble.com. -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
