(Fwd) RE: performance [structured denormalization]

Apparently ODTUG is interested in having someone do a presentation on denormalization (see below).
regards,
ep

• Forwarded message follows ------- Date sent: Thu, 05 Oct 2000 08:55:29 -0800 To: Multiple recipients of list ODTUG-SQLPLUS-L <ODTUG-SQLPLUS-L_at_fatcity.com> Send reply to: ODTUG-SQLPLUS-L_at_fatcity.com From: "Mellott, Eric (EMELLOTT)" <EMELLOTT_at_arinc.com> Subject: RE: performance

Maggie,

I found a couple of articles online regarding denormalization:

"Denormalization and the Rules of Reconstruction" by Dick Root http://www.tdan.com/i014ht04.htm

     [ This article by Dick Root states that there has been a missing
       step in the database design process.  By the time the DBA is
       called up to tune the indices of a database it is probably
       already too late.  Mr. Root makes it clear in this article
       that efficiency should be designed into the data structure
       before the data is actually put on disk  ]

and

"Denormalization Guidelines" by Craig S. Mullins http://www.tdan.com/i001fe02.htm

     [ ...Frequently ...  performance needs dictate very quick
       retrieval capability for data stored in relational databases.
       To accomplish this, sometimes the decision is made to
       denormalize the physical implementation. Denormalization is
       the process of putting one fact in numerous places. This
       speeds data retrieval at the expense of data modification.  

       It is not the intention of this article to promote the concept
       of denormalization. Of course, a normalized set of relational
       tables is the optimal environment and should be implemented
       for whenever possible. Yet, in the real world, denormalization
       is sometimes necessary. Denormalization is not necessarily a
       bad decision if implemented wisely.  ... ]

Hope these help.

• Eric Mellott (not the Eric listed below) :-)

> -----Original Message-----
> From: TOMPKINS, MARGARET [mailto:MARGARET.TOMPKINS_at_dfas.mil]
> Sent: Thursday, October 05, 2000 8:16 AM
> To: Multiple recipients of list ODTUG-SQLPLUS-L
> Subject: RE: performance
>
>
> Eric,
> You have made some valid points as has Mark. I don't
> recall the specific article, but if it was written by Ulka,
> I'm sure it was very good. I would love to see another paper
> on structured denormalization presented at ODTUG this June -
> a paper based on the newer features of Oracle 8 and Designer
> 6. Perhaps you might consider writing one.
>
> Maggie Tompkins
> Defense Finance and Accounting Service
> 816-926-1117 and 317-510-2383
> Margaret.Tompkins_at_dfas.mil
>
> -----Original Message-----
> Sent: Tuesday, October 03, 2000 8:45 PM
> To: Multiple recipients of list ODTUG-SQLPLUS-L
>
>
> Maggie & folks,
>
> I don't know how far back the Oracle Magazine web archives go,
> but I think there there was an article on *structured*
> denormalization by Ulka Rogers a number of years ago (Oracle6?).
>
> If I recall correctly, she said that it is (was anyway) almost
> impossible to find a "real" (enterprise?) production database
> that wasn't designed/implemented with structured denormalization.
> It is (was?) required for performance reasons.
>
> Here are two aspects of implementing with structured denormalization:
>
> 1) specific rules about how to attaining a balance between
> performance and the increase in complexity of a structured
> denormalized application.
>
> 2) sign-off by database owners (users) of the maintenance
> costs/complexities involved in using denormalization to increase
> performance.
>
> (hopefully experts will debunk, correct, or expand on these)
>
> regards,
> ep
>
>
> On 2 Oct 2000, at 15:55, TOMPKINS, MARGARET wrote:
>
> Date sent: Mon, 02 Oct 2000 15:55:24 -0800
> To: Multiple recipients of list
> ODTUG-SQLPLUS-L <ODTUG-SQLPLUS-L_at_fatcity.com>
> Send reply to: ODTUG-SQLPLUS-L_at_fatcity.com
>
> > ... I don't think de-normalization has
> > ever been a blank check for ignoring the basic rules for data
> > integrity.
>
> ...
>
> --
> ODTUG 2000 was a great success! Visit our website at
> www.odtug.com for
> photos, highlights and presentation handouts. And plan now for ODTUG
> 2001 (June 24 to 28) at the Sheraton Hotel and Marina in San
Received on Thu Oct 05 2000 - 14:27:42 CDT