| From: | "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com> |
|---|---|
| To: | "James Pang (chaolpan)" <chaolpan(at)cisco(dot)com> |
| Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Jim Mlodgenski <jimmy76(at)gmail(dot)com>, "pgsql-performance(at)lists(dot)postgresql(dot)org" <pgsql-performance(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: alter table xxx set unlogged take long time |
| Date: | 2022-07-26 15:03:33 |
| Message-ID: | CAKFQuwapFJBOdUPs=ZYsVcDiticJ3mDR2NZ2KxrknPoYBO3iaw@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Tue, Jul 26, 2022 at 5:45 AM James Pang (chaolpan) <chaolpan(at)cisco(dot)com>
wrote:
> Without step 3 , copy data take long time. Use wal_level=minimal can
> help make COPY load data without logging ?
>
>
I believe you are referring to:
https://www.postgresql.org/docs/current/populate.html#POPULATE-COPY-FROM
Since the final state of your table will be "logged" relying on the above
optimization is the correct path, if you enable "logged" at the end, even
with wal_level=minimal, you do not benefit from the optimization and thus
your data ends up being written to WAL.
Otherwise, it is overall time that matters, it's no use boasting the COPY
is fast if you end up spending hours waiting for ALTER TABLE at the end.
David J.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Rick Otten | 2022-07-27 12:55:14 | Postgresql 14 partitioning advice |
| Previous Message | James Pang (chaolpan) | 2022-07-26 12:59:15 | RE: alter table xxx set unlogged take long time |