| Lists: | pgsql-performance |
|---|
| From: | Rod Taylor <pg(at)rbt(dot)ca> |
|---|---|
| To: | Postgresql Performance <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Speeding up pg_dump |
| Date: | 2004-12-14 17:36:46 |
| Message-ID: | 1103045806.25186.153.camel@home |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-performance |
Are there any tricks to speeding up pg_dump aside from doing them from a
replicated machine?
I'm using -Fc with no compression.
--
| From: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
|---|---|
| To: | Rod Taylor <pg(at)rbt(dot)ca> |
| Cc: | Postgresql Performance <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Re: Speeding up pg_dump |
| Date: | 2004-12-14 23:11:27 |
| Message-ID: | 1103065887.4037.3762.camel@localhost.localdomain |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-performance |
On Tue, 2004-12-14 at 17:36, Rod Taylor wrote:
> Are there any tricks to speeding up pg_dump aside from doing them from a
> replicated machine?
>
> I'm using -Fc with no compression.
Run a separate pg_dump for larger tables and run them concurrently so
you use more cpu and disk resources.
The lower compression levels are fast and nearly as good (in my testing)
as full compression. Using compression tends to use up the CPU that
would otherwise be wasted since the pg_dump is disk intensive, and then
saves further I/O by reducing the output file size.
--
Best Regards, Simon Riggs