Re: Impact of checkpoint_segments under continual load conditions

Поиск
Список
Период
Сортировка
От Christopher Petrilli
Тема Re: Impact of checkpoint_segments under continual load conditions
Дата
Msg-id 59d991c405071910133e28031a@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Impact of checkpoint_segments under continual load conditions  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: Impact of checkpoint_segments under continual load conditions
Список pgsql-performance
On 7/19/05, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Christopher Petrilli <petrilli@gmail.com> writes:
> > As I'm doing this, I'm noticing something *VERY* disturbing to me:
> > postmaster backend: 20.3% CPU
> > psql frontend: 61.2% CPU
>
> > WTF?  The only thing going through the front end is the COPY command,
> > and it's sent to the backend to read from a file?
>
> Are you sure the backend is reading directly from the file, and not
> through psql?  (\copy, or COPY FROM STDIN, would go through psql.)

The exact command is:

COPY test (columnlist...) FROM '/tmp/loadfile';

> But even so that seems awfully high, considering how little work psql
> has to do compared to the backend.  Has anyone ever profiled psql doing
> this sort of thing?  I know I've spent all my time looking at the
> backend ...

Linux 2.6, ext3, data=writeback

It's flipped now (stil lrunning), and it's 48% postmaster, 36% psql,
but anything more than 1-2% seems absurd.

Chris
--
| Christopher Petrilli
| petrilli@gmail.com

В списке pgsql-performance по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: Impact of checkpoint_segments under continual load conditions
Следующее
От: Oliver Crosby
Дата:
Сообщение: Looking for tips