Re: [COMMITTERS] pgsql: Install some slightly realistic cost estimation

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Mike Rylander <mrylander(at)gmail(dot)com>
Cc: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [COMMITTERS] pgsql: Install some slightly realistic cost estimation
Date: 2005-04-21 14:57:57
Message-ID: 15397.1114095477@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers pgsql-hackers

Mike Rylander <mrylander(at)gmail(dot)com> writes:
> unrelated indexes? Could this bitmapping code be used to combine
> indexes from different tables, say in a large UNION or inherited table
> setup?
> should be true when the implementation is complete.

No, it's only for combining indexes on a single table --- the bitmaps
only store TIDs, which are relative to a particular table. I don't
think there'd be any advantage to extending the technique to handle
multiple tables; the real win comes when you AND or OR overlapping
bitmaps, and there could never be any overlap between two different
tables.

If we had multi-table indexes it'd be a different story of course.

regards, tom lane

In response to

Browse pgsql-committers by date

  From Date Subject
Next Message Bruce Momjian 2005-04-21 15:20:08 Re: [COMMITTERS] pgsql: Install some slightly realistic cost
Previous Message Tom Lane 2005-04-21 14:49:41 Re: [COMMITTERS] pgsql: Install some slightly realistic cost estimation

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Dunstan 2005-04-21 15:08:48 Re: Postgres: pg_hba.conf, md5, pg_shadow, encrypted passwords
Previous Message Tom Lane 2005-04-21 14:49:41 Re: [COMMITTERS] pgsql: Install some slightly realistic cost estimation