Tom Lane wrote:
> Alvaro Herrera <alvherre@commandprompt.com> writes:
> > While skimming over the pgbench source it has looked to me like it's
> > necessary to pass the -s switch (scale factor) to both the
> > initialization (-i) and the subsequent (non -i) runs.
>
> No, it's not supposed to be, and I've never found it needed in practice.
> The code seems able to pull the scale out of the database (I forget how
> it figures it out exactly).
pgbench is designed to be a general benchmark, meanining it exercises
all parts of the system. I am thinking just reexecuting a single SELECT
over and over again would be a better test of the CPU optimizations.
--
Bruce Momjian bruce@momjian.us
EnterpriseDB http://www.enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +