Re: BUG #4079: libpq.dll very slow (unusable)

Поиск
Список
Период
Сортировка
От Pavel Golub
Тема Re: BUG #4079: libpq.dll very slow (unusable)
Дата
Msg-id 171232314.20080403121304@gf.microolap.com
обсуждение исходный текст
Ответ на Re: BUG #4079: libpq.dll very slow (unusable)  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: BUG #4079: libpq.dll very slow (unusable)  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-bugs
Hello, Tom.

Thank a lot Tom.

Just curious, what was the reason?

Regards

You wrote:

TL> "Pavel Golub" <pavel@microolap.com> writes:
>> Comparing 8.3.1 to 8.2.5 release INSERT of >1000 rows is very slow.

>> INSERT INTO numbers (numberint, numberstr) VALUES (1, 'One');
>> INSERT INTO numbers (numberint, numberstr) VALUES (2, 'Two');
>> INSERT INTO numbers (numberint, numberstr) VALUES (3, 'Three');
>> ...
>> INSERT INTO numbers (numberint, numberstr) VALUES (1000, 'One thousand');

TL> Fixed for 8.3.2.  In the meantime, you might consider modifying your
TL> application to do this as a single insert instead:

TL> INSERT INTO numbers (numberint, numberstr) VALUES
TL>    (1, 'One'),
TL>    (2, 'Two'),
TL>    (3, 'Three'),
TL>    ...
TL>    (1000, 'One thousand')

TL> A quick test says that this way is between three and four times faster,
TL> even without considering the impact of the problem you're complaining
TL> about.

TL>                         regards, tom lane


--
With best wishes,
 Pavel                          mailto:pavel@gf.microolap.com

В списке pgsql-bugs по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: BUG #4085: No implicit cast after coalesce
Следующее
От: "jitendra"
Дата:
Сообщение: BUG #4087: table creation problem using python