| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Mailing Lists -> Oracle-L -> RE: Normalization
I find it very odd that the person who stated this did it. I have a =
pile of respect for Don & find it rather impossible that he made this =
type of statement. Guess we all have things to learn. Anyway, I =
disagree with him, with the exception of a data warehouse. =
Normalization is not to for the purpose of saving disk space, but so =
that when something gets updated you have one and only one place to =
update. If you want speed and normalization use Mineralized views & =
query rewrite. That's what they were made for.
In a data warehouse the purpose is much different in that you want to = preserve history. Therefore renormalizing the data is not just a way = around speed, but the way to preserve the history. Totally different = things altogether and speed or disk space has nothing to do with it.
Dick Goulet
Senior Oracle DBA
Oracle Certified 8i DBA
-----Original Message-----
From: Peter Robson [mailto:pgro_at_bgs.ac.uk]
Sent: Thursday, July 29, 2004 11:51 AM
To: Paul Baumgartel
Subject: Re: Normalization
Outrageous! And this person TAUGHT???
In line...
PB> Q: What are the basic guidelines one should keep in mind while PB> designing a database? Is denormalization always good?
PB> A: I used to teach database design in graduate school, and your
PB> question is excellent! The main purpose of high normalization was =
the
PB> reduction of disk space, back when it mattered in the 1970s.
Absolute utter rubbish - doesn't understand what normalisation is. Doesn't understand what a 'data model' is.
PB> Today, I always introduce redundancy into the model whenever it can
PB> eliminate an SQL join, but not always. I make my decision based on =
two
PB> criteria:
PB> 1 - The size of the redundant item
PB> 2 - The volatility of the item (e.g. how often do I need to =
duplicate
PB> updates)
Oh please - cannot cope with any more of this nonsense....
Hmmm, it might be interesting to know who it was - then perhaps ignorance is bliss...
peter
edinburgh
PB> =3D=3D=3D=3D=3D
PB> Paul Baumgartel
PB> Transcentive, Inc.
PB> www.transcentive.com
=20
PB> __________________________________
PB> Do you Yahoo!?
PB> Yahoo! Mail - You care about security. So do we.
PB> http://promotions.yahoo.com/new_mail
PB> ----------------------------------------------------------------
PB> Please see the official ORACLE-L FAQ: http://www.orafaq.com
PB> ----------------------------------------------------------------
PB> To unsubscribe send email to: oracle-l-request_at_freelists.org
PB> put 'unsubscribe' in the subject line.
PB> --
PB> Archives are at http://www.freelists.org/archives/oracle-l/
PB> FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
PB> -----------------------------------------------------------------
--=20
mailto:pgro_at_bgs.ac.uk
BGS. . http://www.bgs.ac.uk *********************************************************************Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.htmlput 'unsubscribe' in the subject line.
-----------------------------------------------------------------
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request_at_freelists.org
 |
 |