From: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
---|---|
To: | Robert Treat <xzilla(at)users(dot)sourceforge(dot)net> |
Cc: | Josh Berkus <josh(at)agliodbs(dot)com>, pgsql-advocacy(at)postgresql(dot)org |
Subject: | Re: Next version: putting the pedal down |
Date: | 2005-09-30 20:36:49 |
Message-ID: | 1128112609.4045.14.camel@localhost.localdomain |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-advocacy |
On Fri, 2005-09-30 at 09:42 -0400, Robert Treat wrote:
> I like how you separated the two concepts here, but didn't like how
> titled Lance... how about:
Well, I think Josh or Lance should step in here. You're right, I only
meant to separate the two concepts.
> > ! Bitmap Scan: indexes can be dynamically converted to
> > bitmaps in memory, giving up to 20x faster index performance on
> > ! complex queries against against very large tables. Bitmap Scan
> > ! also greatly reduces the need for multi-column indexes.
>
> I think we need to be a little more explicit about why this is a good
> thing.
> Bitmap Scan: indexes can be dynamically converted to
> bitmaps in memory, giving up to 20x faster index performance on
> complex queries against against very large tables. This helps simplify
> database management by greatly reducing the need for multi-column
> indexes.
...This can also simplify database management by greatly reducing
the number of multi-column indexes required for an application.
> > ! Table Partitioning: in version 8.1, the query planner is now
> > ! able to avoid scanning whole sections of a large table without
> > ! needing to utilise indexes using a technique known as
> > ! Constraint Exclusion.
> >
I'll think of a short para.
Best Regards, Simon Riggs
From | Date | Subject | |
---|---|---|---|
Next Message | Lance Obermeyer | 2005-09-30 21:27:06 | Re: PR: release text updated |
Previous Message | Simon Riggs | 2005-09-30 20:29:35 | Re: XA |