Re: Insert Performance

Tom Lane wrote:

> "Michael Paesold" <mpaesold(at)gmx(dot)at> writes:
> > To insert another 10562 rows takes about 12 minutes now!!!
>
> See
> http://www.ca.postgresql.org/users-lounge/docs/7.2/postgres/populate.html
> particularly the point about not committing each INSERT as a separate
> transaction.
>
> regards, tom lane

As I said I wrote a function to insert the rows (PL/pgSQL). All values were
inserted inside a single function call; I always though that a function call
would be executed inside a transaction block. Experience says it does.

About the other points in the docs:

> Use COPY FROM:
Well, I am currently comparing INSERT to COPY ... ;)

> Remove Indexes:
Doesn't COPY also have to update indexes?

> ANALYZE Afterwards:
I have done a VACUUM FULL; VACUUM ANALYZE; just before running the test.

So is it just the planner/optimizer/etc. costs? Would a PREPARE in 7.3 help?

Best Regards,
Michael Paesold