From: | Greg Stark <gsstark(at)mit(dot)edu> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Postgres + Xapian (was Re: fulltext searching via a |
Date: | 2004-01-13 04:32:15 |
Message-ID: | 87zncsqxio.fsf@stark.xeocode.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general pgsql-hackers |
Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Hannu Krosing wrote:
> > Eric Ridge kirjutas R, 09.01.2004 kell 01:16:
> >
> > > Forgetting about composite indexes for a moment, is postgres even
> > > capable of doing 2 index scans in this situation? Does it know how do
> > > take the intersection of two scans?
> >
> > Not currently, but it has been in TODO for quite some time:
> >
> > * Use bitmaps to fetch heap pages in sequential order [performance]
> > * Use bitmaps to combine existing indexes [performance]
>
> And this the foundation of bitmapped indexes, as I understand it, and is
> used for "star" joins, popular with data warehousing.
Actually I think the thinking with those two TODO items is to use these
algorithms in conjunction with regular indexes. Doing regular index scans in
parallel, construct bitmaps, perform logical operations to join them, and then
do the table scan.
It could be done in conjunction with a scheme for storing bitmap indexes or
not. Probably not though.
--
greg
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2004-01-13 04:43:49 | Re: Vacuum Error |
Previous Message | Alex Satrapa | 2004-01-13 03:53:16 | Re: sql insert function |
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2004-01-13 04:46:40 | Re: LWLock/ShmemIndex startup question |
Previous Message | Claudio Natoli | 2004-01-13 00:11:29 | Re: LWLock/ShmemIndex startup question |