Message-ID: <33039.67.87.27.161.1102605925.squirrel@www.ke5in.com>
Date: Thu, 9 Dec 2004 10:25:25 -0500 (EST)
Subject: filtering
From: "Kevin B." <db@ke5in.com>
To: pgsql-sql@postgresql.org
Reply-To: db@ke5in.com
User-Agent: SquirrelMail/1.4.2
MIME-Version: 1.0
Content-Type: text/plain;charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Importance: Normal

Hi,

I have a 14 million row table with one index on two fields one is a varchar
the other is a date.  The combination of the two makes the row unique.

Data
-----------------
name  date    ... other fields
a     1/1/01
a     1/2/01
a     1/3/01
b     1/1/01
b     1/2/01
d     1/1/01
d     1/2/01

I have a table with just the names.  each name occurs once.
UName
---------
name
a
b
c
d

I've tried a number of queries to find which name is in UName but not in
Data.   However, they are all taking too long (more than 30 minutes - but
the hard drive is a slow 4200rpm IDE....).

What is the quickest query to get the result that I want?  Also, should I
put another index on the Data table for "name" only?

Thanks