| From: | Niels Jespersen <NJN(at)dst(dot)dk> |
|---|---|
| To: | "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Possibilities for optimizing inserts across oracle_fdw foreign data wrapper |
| Date: | 2021-09-19 10:28:48 |
| Message-ID: | ae9ab150af5a417e985e581249107853@dst.dk |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hello all
We are often using the oracle_fdw to transfer data between Postgres (version 11+) and Oracle (version 18+). It works great.
However I have a task at hand that requires inserting a few billion rows in an Oracle table from a Postgres query.
insert into t_ora (a,b,c)
select a,b,c from t_pg;
This is driven from a plpgsql stored procedure, if that matters.
I want to optimize the running time of this. But I am unsure of which, if any, possibilities there actually is.
Reducing the number of network roundtrips is usually a good way to increase throughput. But, how do I do that?
If I could make the Oracle insert direct load, that would usually also increase throughput. But, is that possible here. There are no constraints defined on the destinaton tables.
Regards Niels Jespersen
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Mladen Gogala | 2021-09-20 01:52:00 | Re: Possibilities for optimizing inserts across oracle_fdw foreign data wrapper |
| Previous Message | Tobias Meyer | 2021-09-18 19:00:18 | Re: Remove duplicated row in pg_largeobject_metadata |