From: | Alexander Staubo <alex(at)purefiction(dot)net> |
---|---|
To: | "Brandon Aiken" <BAiken(at)winemantech(dot)com> |
Cc: | "Tomi NA" <hefest(at)gmail(dot)com>, <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: performace review |
Date: | 2006-10-07 21:59:59 |
Message-ID: | A94CC970-36F0-4BD2-A37D-DC4777CEDF8F@purefiction.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Oct 7, 2006, at 23:44 , Brandon Aiken wrote:
> I download the db source (inside opencrx-1.9.1-
> core.postgresql-8.zip) and executed their three schema files,
> dbcreate-indexes.sql, dbcreate-views.sql, dbcreate-tables.sql.
> Each of the 118 tables has a three-field composite primary key of
> 'PRIMARY KEY (object_rid, object_oid, object_idx)'. object_rid and
> object_oid are both VARCHAR(200). There are *no* foreign key
> constraints. Each table has between 15 and 50 fields, with 25
> looking about average.
To be fair, there are a bunch of indexes, but the number of indexes
seems low compared to the number of fields.
> Gee, why to table joins take so long? Maybe because a blind monkey
> created the schema? Normalized databases do tend to perform
> better, so I hear.
*De*normalization is the traditional hack to speed up queries,
because it reduces the need for joins.
Alexander.
From | Date | Subject | |
---|---|---|---|
Next Message | Jorge Godoy | 2006-10-07 22:07:47 | Re: increment row number function question |
Previous Message | Brandon Aiken | 2006-10-07 21:44:02 | Re: performace review |