Re: PostgreSQL Write Performance

Поиск
Список
Период
Сортировка
От Tim Uckun
Тема Re: PostgreSQL Write Performance
Дата
Msg-id 855e4dcf1001051546n2f4a481erd97f582fa12e8445@mail.gmail.com
обсуждение исходный текст
Ответ на Re: PostgreSQL Write Performance  (Craig Ringer <craig@postnewspapers.com.au>)
Ответы Re: PostgreSQL Write Performance  (Christophe Pettus <xof@thebuild.com>)
Re: PostgreSQL Write Performance  (Craig Ringer <craig@postnewspapers.com.au>)
Список pgsql-general
>
> Technically you *can* disable triggers, including RI checks, but it's VERY
> unwise and almost completely defeats the purpose of having the checks. In
> most such situations you're much better off dropping the constraints then
> adding them again at the end of the load.


I know that the SQL server bulk loader defaults to not firing the
triggers and I was hoping there would be an option on the COPY command
to accomplish the same thing.

pg_dump has a --disable-triggers option too.

It seems to me that the COPY FROM  should have an option that bypasses
the triggers as a convience.

Both the SQL server team and the postgres team have obviously
recognized that there will be situations where the DBA will want to
bulk load data without firing off a set of triggers for every insert.

It doesn't seem like an outrageous expectation that the COPY command
or something similar should have that option.

В списке pgsql-general по дате отправления:

Предыдущее
От: Roman Neuhauser
Дата:
Сообщение: Re: set-level update fails with unique constraint violation
Следующее
От: Greg Smith
Дата:
Сообщение: Re: pgbench out of memory error