Re: Impact of checkpoint_segments under continual load conditions

On 7/19/05, Christopher Petrilli <petrilli(at)gmail(dot)com> wrote:
> On 7/19/05, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > Christopher Petrilli <petrilli(at)gmail(dot)com> writes:
> > > On 7/19/05, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > >> How *exactly* are you invoking psql?
> >
> > > It is a subprocess of a Python process, driven using a pexpect
> > > interchange. I send the COPY command, then wait for the '=#' to come
> > > back.
> >
> > Some weird interaction with pexpect maybe? Try adding "-n" (disable
> > readline) to the psql command switches.
>
> Um... WOW!
> It doesn't stay QUITE that low, but it stays lower... quite a bit.
> We'll see what happens over time.

here's a look at the difference:

http://blog.amber.org/diagrams/pgsql_readline_impact.png

I'm running additional comparisons AFTER clustering and analyzing the tables...

Chris
--
| Christopher Petrilli
| petrilli(at)gmail(dot)com