Neil Conway wrote:
> On Thu, 20 Jun 2002 22:50:04 -0400 (EDT)
> "Bruce Momjian" <pgman@candle.pha.pa.us> wrote:
> > I have thought that some type of feedback from the executor back into
> > the optimizer would be a good feature. Not sure how to do it, but your
> > idea makes sense. It certainly could update the table statistics after
> > a sequential scan.
>
> Search the archives for a thread I started on -hackers called "self-tuning
> histograms", which talks about a pretty similar idea. The technique there
> applies only to histograms, and builds the histogram based *only* upon
> the data provided by the executor.
>
> Tom commented that it's probably a better idea to concentrate on more
> elementary techniques, like multi-dimensional histograms, before starting
> on ST histograms. I agree, and plan to look at multi-dimensional histograms
> when I get some spare time.
I was thinking of something much more elementary, like a table that
reports to have 50 blocks but an executor sequential scan shows 500
blocks.
-- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610)
853-3000+ If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill,
Pennsylvania19026
