From: | Dennis Gearon <gearond(at)fireserve(dot)net> |
---|---|
To: | Bruno Wolff III <bruno(at)wolff(dot)to> |
Cc: | Michael Fuhr <mike(at)fuhr(dot)org>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: flattened tables with normalized tables |
Date: | 2004-10-08 05:12:07 |
Message-ID: | 416621A7.9040208@fireserve.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Bruno Wolff III wrote:
> On Thu, Oct 07, 2004 at 22:35:50 -0600,
> Michael Fuhr <mike(at)fuhr(dot)org> wrote:
>
>>On Thu, Oct 07, 2004 at 09:08:18PM -0700, Dennis Gearon wrote:
>>
>>>About regular views, how does that speed things up, other than the initial
>>>SQL interpretation of the view not needing to be done?
>>
>>I didn't mean to imply that views would speed things up -- I was
>>merely suggesting them as an alternative to your "flattened table"
>>if part of its purpose would be to simplify queries. You might
>>want to perform some experiments to see if the performance gains
>>from a materialized view are worth the extra complexity.
>
>
> It may even turn out there aren't any performance gains from having a
> materialized view. That will depend on the mix of operations in production.
>
Well, one particular query / view will probably draw from 11-15 tables. Several of those tables should have millions and millions of rows. However,as normalzed as all the data is, and having used surrogate, integer primary keys, the tables shouldn't be that big, most of them.
From | Date | Subject | |
---|---|---|---|
Next Message | Bruno Wolff III | 2004-10-08 05:17:27 | Re: flattened tables with normalized tables |
Previous Message | Dennis Gearon | 2004-10-08 04:44:00 | Re: flattened tables with normalized tables |