| From: | "Scott Marlowe" <smarlowe(at)qwest(dot)net> |
|---|---|
| To: | "Simon Windsor" <simon(dot)windsor(at)cornfield(dot)org(dot)uk> |
| Cc: | "Postgres List" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Postgres tuning? |
| Date: | 2004-06-29 22:33:48 |
| Message-ID: | 1088548428.12350.41.camel@localhost.localdomain |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Tue, 2004-06-29 at 15:30, Simon Windsor wrote:
> Hi
>
> I am in the process of converting a small multi-user application from
> MySQL, and most queries are performing better. The only noticeable
> exception is a batch load, which is half the speed of MySQL version.
>
> What are the basic parameters I should be focusing on for best
> performance ?
>
> - sort_mem
> - shared_buffers
How does the batch insert work? Just one insert after another? Try
wrapping them in a transaction:
begin;
insert into ... (repeat a few thousand to million times)
commit;
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Paul Thomas | 2004-06-29 23:01:47 | Re: Postgres tuning? |
| Previous Message | Stan Y | 2004-06-29 22:09:19 | What is the prevalent naming convention in RDBMS? |