Re: Slow delete
От | Tom Lane |
---|---|
Тема | Re: Slow delete |
Дата | |
Msg-id | 20073.1121276768@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Slow delete (Doug Hall <doughalldev@gmail.com>) |
Ответы |
Re: Slow delete
|
Список | pgsql-general |
Doug Hall <doughalldev@gmail.com> writes: > delete from citizen where id not in (select citizenid from > citizen_stage); > The explain select tells me that there is a sequential select of > citizen_stage records. (??) There are 75009 citizen records and 14778 > records, and it's taking more than half an hour. How can I speed this > up? How old is your Postgres? I'd expect 7.4 and up to do this with a hashed IN, which'd be reasonably fast. If the EXPLAIN output doesn't say anything about a "hashed subplan", then either you've got an old version or there's some sort of estimation problem. If it is a hashed IN and it's still slow, I'd wonder about unindexed foreign key references to the citizen table. regards, tom lane
В списке pgsql-general по дате отправления: