Re: Impact of checkpoint_segments under continual load conditions - Mailing list pgsql-performance

Christopher Petrilli <petrilli@gmail.com> writes:
>> Are you sure the backend is reading directly from the file, and not
>> through psql?  (\copy, or COPY FROM STDIN, would go through psql.)

> The exact command is:
> COPY test (columnlist...) FROM '/tmp/loadfile';

I tried to replicate this by putting a ton of COPY commands like that
into a file and doing "psql -f file ...".  I don't see more than about
0.3% CPU going to psql.  So there's something funny about your test
conditions.  How *exactly* are you invoking psql?

            regards, tom lane