--On Donnerstag, Februar 12, 2009 16:06:31 -0800 "Joshua D. Drake"
<jd(at)commandprompt(dot)com> wrote:
> However, in recent times I have found that increasing cpu_tuple_cost,
> cpu_operator_cost and cpu_index_tuple_cost to be very useful. This is
> always in the scenario of, "queries were running fine for months and
> then all of a sudden, they are not". It is also always on systems that
> we are already maintaining and thus (in theory) are in good shape.
Hmm have you tried seq_page_cost and random_page_cost, too? I found them
really important especially if you have a steadily growing database or a
fully cached database to reflect the "real" disk access costs.
--
Thanks
Bernd