Re: Limit + group + join
От | Mark Kirkwood |
---|---|
Тема | Re: Limit + group + join |
Дата | |
Msg-id | 431109FB.6000505@paradise.net.nz обсуждение исходный текст |
Ответ на | Re: Limit + group + join (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: Limit + group + join
|
Список | pgsql-performance |
Tom Lane wrote: > > I looked into this and found that indeed the desirable join plan was > getting generated, but it wasn't picked because query_planner didn't > have an accurate idea of how much of the join needed to be scanned to > satisfy the GROUP BY step. I've committed some changes that hopefully > will let 8.1 be smarter about GROUP BY ... LIMIT queries. > Very nice :-) joinlimit=# EXPLAIN SELECT c.id FROM c JOIN b ON c_id=c.id GROUP BY c.id ORDER BY c.id DESC LIMIT 5; QUERY PLAN -------------------------------------------------------------------------------------------------- Limit (cost=0.00..15.23 rows=5 width=4) -> Group (cost=0.00..243730.00 rows=80000 width=4) -> Nested Loop (cost=0.00..243530.00 rows=80000 width=4) -> Index Scan Backward using c_pkey on c (cost=0.00..1450.00 rows=80000 width=4) -> Index Scan using b_on_c on b (cost=0.00..3.01 rows=1 width=4) Index Cond: (b.c_id = "outer".id) (6 rows) This is 8.1devel from today. regards Mark
В списке pgsql-performance по дате отправления: