Re: Test request for Stats collector performance improvement

Qingqing Zhou wrote:
>
> "Bruce Momjian" <pgman(at)candle(dot)pha(dot)pa(dot)us> wrote
> >
> > > -- After patch --
> > >
> > > real 0m1.275s
> > > user 0m0.097s
> > > sys 0m0.160s
> > >
> > > real 0m4.063s
> > > user 0m0.663s
> > > sys 0m0.377s
> > >
> > > real 0m1.259s
> > > user 0m0.073s
> > > sys 0m0.160s
> >
> > I assume the above is just running the same test three times, right?
> >
>
> Right -- it is the result of the patched CVS tip runing three times with
> stats_command_string = on. And the tests marked "--Before patch--" is the
> result of CVS tip running three times with stats_command_string = on.

Any idea why there is such a variance in the result? The second run
looks quite slow.

--
Bruce Momjian http://candle.pha.pa.us
EnterpriseDB http://www.enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +