tgl@sss.pgh.pa.us (Tom Lane) wrote in message
> For the moment, anyway. There have been a couple of rounds of
> pgsql-hackers discussion about whether to lower the default value of
> random_page_cost, but so far no one has done any experiments that
> would be needed to establish a good new value. (The current default
> of 4.0 is based on some old experiments I did. I'm quite willing to
> accept that those experiments might have been flawed, but not willing
> to replace the number without seeing better experiments...)
>
I wonder if a simple script could be designed to test the various
parameters specified in postgresql.conf. The user could be asked to
input 3 or 4 of the most commmon database queries and the script would
determine the cost, run those queries, and send back the actual time
of completion for each variation of the optimizer (e.g. run 1,
enable_seq_scan false; run 2, random_page_cost 3; etc.).
This way administrators would have a way (albeit brute force) to
fine-tune their settings to their specific machine and data. Plus,
they could upload the results to the hackers list just like regression
tests.
Just a thought.
-Tony