Merge Data Sets

Presently i've created a numeric key on each table and joined that way but it means I have to use a sequence/update combo i am just looking for anyone that is aware of a faster approach...

-----Original Message-----
From: Galen Boyer [mailto:galenboyer_at_hotpop.com] Posted At: Tuesday, September 25, 2001 7:50 PM Posted To: server
Conversation: Merge Data Sets
Subject: Re: Merge Data Sets

On Tue, 25 Sep 2001, AJohnson_at_Armature.com wrote:

> Hi, Does anyone have any experience with merging two data sets
> that have no relationship whatsoever.

I don't other than to create a _really_ large table from smaller tables for some sort of testing area.

> for example.
>
> Table X ( Acol varchar2,Bcol varchar2,dob date)
>
> Table Y (XCol Number)
>
> |
> |
>
> Table Z (XCol number, Acol varchar2,Bcol varchar2,dob date)
>
> these table have no join key and thier is around 25 million
> rows..

I assume you understand the term cartesian product? If both tables have 25 rows, then you will create 25 x 25 = 625 rows. Therefore, your cartesian product will create a 625 million row table.

> any help would be appreciate, I have written a bit of pl/sql
> but it is miserably slow, i am wanting to take advantage of
> direct insert hint.

A 625 million row insert could probably take awhile. Are you needing to optimize this insert, or were you expecting a different result from your merging based on some other criteria?

-- 
Galen Boyer
Everyone in town, now, they probably all agree,
I'm _lying_ in the bed I made.