hmm, how 100 000 page buffer can make me run out of physical memory ? with 4ko page buffer, it's only 400 mb of memory and i have 4gb on our test machine.

but anyway that was a good remarque (and a good guess) and it's look logical. how to know if a program go into a memoy disk swap hell ?

thanks again ann !
stephane

--- In firebird-support@yahoogroups.com, "Ann W. Harrison" <aharrison@...> wrote:
>
> svanderclock wrote:
> > no one have an idea of this behavior ?
> >> now i encountered another strange behavior, it's when i setup a big page buffer
> >>
> >> page buffer 2048: execute time 156ms
> >> page buffer 20480: execute time 156ms
> >> page buffer 100000: execute time 15288 ms !!!
> >>
> >> any idea why when i setup a big page buffer then the execute time go down by 100 ! the commit time don't change..
> >>
> >> many thanks by advance for your precious help !!
>
> My guess is that at 100,000 page buffers, you've run out
> of physical memory and are page faulting. Tuning a database
> is a little like cooking - the fact that a bit of salt improves
> the flavor doesn't imply that a kilo of salt will make it
> delicious.
>
> Cheers,
>
> Ann
>