Re: performance in selecting from very large recordset - Mailing list pgsql-general

From Bruno Wolff III
Subject Re: performance in selecting from very large recordset
Date
Msg-id 20031219213346.GC14968@wolff.to
Whole thread Raw
In response to Re: performance in selecting from very large recordset  ("Dann Corbit" <DCorbit@connx.com>)
List pgsql-general
On Fri, Dec 19, 2003 at 12:54:35 -0800,
  Dann Corbit <DCorbit@connx.com> wrote:
>
> There were problems with hashed indexes in PostgreSQL before.  I think
> that they have all been addressed in 7.4, but I am not an expert on
> their usage in that environment, so it would be good if others can
> supply feedback.  I develop database systems under many environments,
> and so my advice is not PG specific.

While some bugs with hashed indexes were fixed, as far as I know they
still aren't faster than btree indexes for anything. So that you would
want to use btree indexes in preference to hash indexes.

pgsql-general by date:

Previous
From: Bruno Wolff III
Date:
Subject: Re: Tables referencing each other
Next
From: Jonathan Bartlett
Date:
Subject: Re: Salvage older PostgreSQL data disk - help?