Re: Hash index use presently(?) discouraged since 2005: revive or bury it?

Поиск
Список
Период
Сортировка
От Claudio Freire
Тема Re: Hash index use presently(?) discouraged since 2005: revive or bury it?
Дата
Msg-id CAGTBQpanG5vXJctZ42Qz9F9=+pqQjrYyjM6emZio6brh4xQELw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Hash index use presently(?) discouraged since 2005: revive or bury it?  (Merlin Moncure <mmoncure@gmail.com>)
Список pgsql-performance
Дерево обсуждения 
On Fri, Sep 16, 2011 at 3:00 AM, Merlin Moncure <mmoncure@gmail.com> wrote:
>
> c:\Program Files\PostgreSQL\9.0\data>dir/s | grep 16525
> 09/15/2011  07:46 PM       224,641,024 16525
>
> c:\Program Files\PostgreSQL\9.0\data>dir/s | grep 16526
> 09/15/2011  07:49 PM       268,451,840 16526

That's not surprising at all.
Hashes need to be bigger to avoid collisions.

What's more interesting than index creation, is index maintainance and
access costs.
In my experience, btree beats hash.
I haven't tried with 9.1, though.

В списке pgsql-performance по дате отправления:

Предыдущее
От: Merlin Moncure
Дата:
Сообщение: Re: Hash index use presently(?) discouraged since 2005: revive or bury it?
Следующее
От: Merlin Moncure
Дата:
Сообщение: Re: Hash index use presently(?) discouraged since 2005: revive or bury it?