Re: Restore v. Running COPY/INDEX seperatly

Поиск
Список
Период
Сортировка
От Benjamin Arai
Тема Re: Restore v. Running COPY/INDEX seperatly
Дата
Msg-id 48C98F54-E517-4208-80C7-D92856C55190@benjaminarai.com
обсуждение исходный текст
Ответ на Re: Restore v. Running COPY/INDEX seperatly  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-general
In what order should I :

- COPY data
- Create indexes
- Create Trigger
- Vaccum

?

Currently I am:

1. Create table
2 . Create trigger for updates
3. Create indexes including gin
4. Vaccum

Benjamin

On Aug 27, 2007, at 7:59 AM, Tom Lane wrote:

> Benjamin Arai <me@benjaminarai.com> writes:
>> Why is a trigger faster than doing a ALTER after table is created?  I
>> thought a trigger would be slower because it would be invoked every
>> iteration (a new row is inserted) during the COPY process.
>
> Yeah, you'd have the trigger overhead, but the above argument ignores
> the costs of the full-table UPDATE --- not to mention the VACUUM
> you'll need after the UPDATE to clean up the dead rows.
>
>             regards, tom lane
>


В списке pgsql-general по дате отправления:

Предыдущее
От: "Carlo Stonebanks"
Дата:
Сообщение: Re: Will UPDATE lock if FROM refers to target table?
Следующее
От: Lee Keel
Дата:
Сообщение: Re: Convert bytea to Float8