BITMAP Index support (and other DSS info.)

From: "Shahbaz Chaudhary" <chaudhar(at)umich(dot)edu>
To: <pgsql-hackers(at)postgresql(dot)org>
Subject: BITMAP Index support (and other DSS info.)
Date: 2002-12-29 21:44:09
Message-ID: 000001c2af83$6b98fc80$6401a8c0@shahbaz
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi all,
I've recently started using Postgresql and am impressed with how
complete an RDBMS system it really is. It seems that while some
corporations may balk at using it for day to day operational use,
bringing in PGSQL for a more niche use will allow many professionals to
see its potential.

One such niche market is decision support systems (DSS). Doing
something like Sybase IQ on PGSQL (bitmap/bitwise index) would greatly
help exposure. I noticed that adding BITMAP index is not only on the
TODO list but has also been discussed quite a bit in the past. I'm not
sure if any one has seen this
http://www.it.iitb.ernet.in/~rvijay/dbms/proj/ , a bitmap index
implementation using GiST! I, my self, run a DB (about 2 GIG and
growing) with over ten million rows...and the need to be responsive
enough to be used on the web. This would be great.

I haven't seen this in any of the mailing lists, a new book "PostgreSQL"
by Korry Douglas (comes out in februrary '03)
(http://www.amazon.com/exec/obidos/tg/detail/-/0735712573/qid=1041197382
/sr=1-8/ref=sr_1_8/104-4850283-3895915?v=glance&s=books) which explains
the internals of PostgreSQL ...good for people who would like to
contribute to this project but haven't done hardcore C/C++ programming
since college (like myself).

There are bound to be people in the academia (grad students, professors
of CS, etc.) on this mailing list, yet I see few RDBMS courses using
postgresql as an example. If people still have connections to
universities, it would seem that inviting researchers to use PGSQL for
their experiments will quickly make it comparable to Oracle/etc. This
would be specifically helpful for adding capabilities that are not
considered top priority (time-series DBs...I work in a fiancial firm and
trust me when I say that they spend HUGE amounts of money on software
that does nothing more than what postgresql developers could add:
ordered set of tuples, link to data streams, memory based tables, etc.)

Any way, I hope this isn't considered off-topic :).

Shahbaz C.

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Kevin Brown 2002-12-30 00:39:37 Re: MOVE strangeness
Previous Message Bruce Momjian 2002-12-29 04:36:02 IPv6 patch