Re: Subject: bool / vacuum full bug followup part 2 - Mailing list pgsql-general

From Tom Lane
Subject Re: Subject: bool / vacuum full bug followup part 2
Date
Msg-id 15976.1020703443@sss.pgh.pa.us
Whole thread Raw
In response to Re: Subject: bool / vacuum full bug followup part 2  (Scott Marlowe <scott.marlowe@ihs.com>)
List pgsql-general
Scott Marlowe <scott.marlowe@ihs.com> writes:
> Since postgresql 7.2.1 refuses to make an index of any kind other than
> btree, what is the answer?  no indexes?  While single column indexes may
> seem wasteful, remember that the boole may be stored in a table that has
> very large tuples, and a sequential scan of such a table could be quite
> slow, or there may be a situation where a tiny percentage of the booles
> are one setting while most are the other, like an approval system for
> online content.

You could use a partial index for the latter case.  If the column is
about fifty-fifty then any kind of index is a waste of space...

            regards, tom lane

pgsql-general by date:

Previous
From: "Nigel J. Andrews"
Date:
Subject: Re: pgaccess
Next
From: "Johnson, Shaunn"
Date:
Subject: pg_restore question