Re: PostgreSQL pre-fork speedup

Поиск
Список
Период
Сортировка
Искать
От
James Robinson
Тема
Re: PostgreSQL pre-fork speedup
Дата
в 16:07:25
Msg-id
8B081EC4-9F90-11D8-9C07-000A9566A412@socialserve.com
Ответ на
Список
Дерево обсуждения
PostgreSQL pre-fork speedup sdv mailer <sdvmailer@yahoo.com>
Re: PostgreSQL pre-fork speedup Paul Ramsey <pramsey@refractions.net>
Re: PostgreSQL pre-fork speedup Tom Lane <tgl@sss.pgh.pa.us>
Re: PostgreSQL pre-fork speedup sdv mailer <sdvmailer@yahoo.com>
Re: PostgreSQL pre-fork speedup Greg Stark <gsstark@mit.edu>
Re: PostgreSQL pre-fork speedup sdv mailer <sdvmailer@yahoo.com>
Re: PostgreSQL pre-fork speedup Andrew Sullivan <ajs@crankycanuck.ca>
Re: PostgreSQL pre-fork speedup sdv mailer <sdvmailer@yahoo.com>
Re: PostgreSQL pre-fork speedup Richard Huxton <dev@archonet.com>
Tom Lane writes:... too much flushing ...

I agree. I'll bet replacing the pool_write_and_flush() calls in 
BinaryRow()
and AsciiRow() with just pool_write(), followed by removing the
fflush() calls at the bottom of those two methods should go a long
way towards fixing things, since the CompleteCommandResponse
handler method ends with a call to pool_write_and_flush(), and
you've pretty much gotta get a CompleteCommand message
trailing all of those rows.

----
James Robinson
Socialserve.com


В списке pgsql-hackers по дате отправления
От: Tom Lane
Дата:
Сообщение: Re: ALTER TABLE TODO items
От: scott.marlowe
Дата:
FAQ