> But to get the estimated cost ratio to match up with the actual cost
> ratio, we'd have to raise random_page_cost to nearly 70, which is a bit
> hard to credit. What was the platform being tested here?
Why ? Numbers for modern single disks are 1-2Mb/s 8k random and 50-120 Mb/s
sequential.
An
> >Incrementing random_page_cost from 4 (the default) to 5 causes the
> >planner to make a better decision.
>
> We have such a low default random_page_cost primarily to mask other
> problems in the optimizer, two of which are
>
> . multi-column index correlation
>
> . interpolation between min_
"Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:
> ... If bitmap
> scan is disabled, the planner will pick index scan even in cases when
> sequential scan is 10x faster:
> scratch=# set enable_bitmapscan to off;
> SET
> scratch=# explain analyze select count(1) from test where random >=
> 142907698
On Tue, 17 May 2005 22:12:17 -0700, "Jeffrey W. Baker" <[EMAIL PROTECTED]>
wrote:
>Incrementing random_page_cost from 4 (the default) to 5 causes the
>planner to make a better decision.
We have such a low default random_page_cost primarily to mask other
problems in the optimizer, two of which are
On Mon, 2005-05-16 at 14:35 -0400, Tom Lane wrote:
> "Jeffrey W. Baker" <[EMAIL PROTECTED]> writes:
> > On Mon, 2005-05-16 at 09:53 -0400, Tom Lane wrote:
> >> This is a fallacy, and I think your concern is largely mistaken. Have
> >> you experimented with the cases you are worried about?
>
> > P
