I was thinking in something independent from the executor, simply a variable
that recommends or not the use of a particular index, it could be obtained
from user, and so it could be improved(a factor lower than 1) on planner.
How about something like this?
----- Original Message -----
From: "Bruce Momjian" <pgman@candle.pha.pa.us>
To: "Luis Alberto Amigo Navarro" <lamigo@atc.unican.es>
Cc: "Lincoln Yeoh" <lyeoh@pop.jaring.my>; "Tom Lane" <tgl@sss.pgh.pa.us>;
"mlw" <markw@mohawksoft.com>; "Andrew Sullivan" <andrew@libertyrms.info>;
"PostgreSQL-development" <pgsql-hackers@postgresql.org>
Sent: Tuesday, April 23, 2002 6:42 PM
Subject: Re: [HACKERS] Index Scans become Seq Scans after VACUUM ANALYSE
> Luis Alberto Amigo Navarro wrote:
> > Hi All.
> > I've been reading all the thread and I want to add a few points:
> >
> > You can set enable_seqscan=off in small or easy queries, but in large
> > queries index can speed parts of the query and slow other, so I think it
is
> > neccesary if you want Postgres to become a Wide-used DBMS that the
planner
> > could be able to decide accuratelly, in the thread there is a point that
> > might be useful, it will be very interesting that the planner could
learn
> > with previous executions, even there could be a warm-up policy to let
> > planner learn about how the DB is working, this info could be stored
with DB
> > data, and could statistically show how use of index or seqscan works on
> > every column of the DB.
>
> Yes, I have always felt it would be good to feed back information from
> the executor to the optimizer to help with later estimates. Of course,
> I never figured out how to do it. :-)
>
> --
> 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, Pennsylvania 19026
>
