Re: What is wrong with hashed index usage?

> -----Original Message-----
> From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
> Sent: Friday, June 21, 2002 1:31 PM
> To: Dann Corbit
> Cc: Tom Lane; Neil Conway; mloftis@wgops.com;
> pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] What is wrong with hashed index usage?
>
>
> Dann Corbit wrote:
> > > I was thinking of this during CREATE INDEX ... hash:
> > >
> > >     NOTICE:  Hash index use is discouraged.  See the CREATE INDEX
> > >     reference page for more information.
> > >
> > > Does anyone else like/dislike that?
> >
> > I think it might be OK temporarily, to show that there is
> some work that
> > needs done.  When hashed indexes are fixed, the notice should be
> > removed.
>
> Oh, yes, clearly, we would remove it once we had a hash implementation
> that had _any_ advantages over btree.
>
> So, is you vote for or against the elog(NOTICE)?

I will defer to the preference of the others.  I lean ever so slightly
towards the notice, because it is very unusual for hashed index not to
be faster for single item lookup.