Re: An optimization question

2007/3/9, Ohad Lutzky <lutzky-Re5JQEeQqe8AvxtiuMwx3w@xxxxxxxxxxxxxxxx>:
> On 3/9/07, Clodoaldo 
> <clodoaldo.pinto.neto-Re5JQEeQqe8AvxtiuMwx3w@xxxxxxxxxxxxxxxx> wrote:
> > 2007/3/9, Ohad Lutzky <lutzky-Re5JQEeQqe8AvxtiuMwx3w@xxxxxxxxxxxxxxxx>:
> > > And when running them,
> > > $ time python benchmark.py && time python benchmark2.py
> > >
> > > real    0m2.876s
> > > user    0m0.624s
> > > sys     0m0.192s
> > >
> > > real    0m1.503s
> > > user    0m0.368s
> > > sys     0m0.196s
> >
> > What happens if you invert the benchmark order?:
> > $ time python benchmark2.py && time python benchmark.py
>
> $ time python benchmark2.py  && time python benchmark.py
>
> real    0m1.526s
> user    0m0.432s
> sys     0m0.104s
>
> real    0m2.777s
> user    0m0.556s
> sys     0m0.224s
>
> No difference.
>
> > I use prepared queries wiht psycopg2 hopping it would lead to better
> > performance, but I have never benchmarked it.
>
> Do you mean by using explicit "PREPARE" and "EXECUTE" queries, or by
> some other means?

I use explicit "PREPARE" and "EXECUTE"

> Because even a manual way to prepare a statement
> (similar to the one available in Perl's DBI, $statement =
> $dbh->prepare("SELECT * FROM table WHERE param=?");
> $statement->execute(5);) would be great.

Regards,
-- 
Clodoaldo Pinto Neto