Re: Perfornamce Q

Поиск
Список
Период
Сортировка
От Jean-Christian Imbeault
Тема Re: Perfornamce Q
Дата
Msg-id 3E2F7F1F.5050504@mega-bucks.co.jp
обсуждение исходный текст
Ответ на Perfornamce Q  (Jean-Christian Imbeault <jc@mega-bucks.co.jp>)
Ответы Re: Perfornamce Q  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-general
Tom Lane wrote:
>
> I think it's the foreign keys that are hurting you.  Are the referenced
> columns indexed?

They should be, they are all primary keys in the referenced tables.

But why would they be hurting me? The update is on a column that has no
constraints on it. If the foreign key constraints are hurting me I don't
understand the reason ... could you explain why?

> Are they of the same datatypes as the referencing
> columns?

They must be no? Otherwise how could they be foreign key constraints?
(And I just checked and yes they are of the same type).

> Have you ANALYZEd those tables lately?

In the output I gave the first thing I did was a "vacuum full analyze".
I have actually found that the time require usually goes *up* strangely
enough after doing a vacuum full analyze though ...

Jc


В списке pgsql-general по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: Perfornamce Q
Следующее
От: Justin Clift
Дата:
Сообщение: Re: [HACKERS] C++ coding assistance request for a visualisation tool