Re: true grits / Re: Where can I find real-life-examples about ORACLE installations?

> > I guess I'm saying that I can't recall starting with a completely
> > normalized database ( just 3rd normal form here ) and then denormalize if
> > we found it necessary for some reason.
> >
> > We've usually have had some denormalization in as soon as we started
> > doing physical modeling. Sigh.
>
> Well, why would you *not* want to denormalize during design? It seems
> to me that (theoretically) ***if*** you are doing "structured"
> denormalization correctly, that is exactly when you would want to do
> it, no?

Unless you detect a performance problem, why denormalize at all?

We always have folks that want to denormalize because they *think* there will be a performance problem. This usually occurs because they think that joining 3 or 4 tables will be too slow.

We're talking strictly OLTP databases here. Joins for OLTP are usually quite fast, and performance problems won't be fixed by denormalizing.

I once worked on a database that had to do a number of rather complex joins in an OLTP database, with a hard limit of 9 seconds for the response time.

That 9 seconds was from the time a pharmacist hit a key on a terminal in the pharmacy. During that 9 seconds the data had to be routed to a clearinghouse in Atlanta Georgia, sent to us, used to adjudicate a claim and send a response back to the pharmacy through the same network.

The average adjudication was 1.5 seconds if I recall correctly, and this was on a fairly normalized database. Only one table was highly denormalized, and that was nobody could figure out a reasonable way to normalize it. Not sure if I could yet. :)

Anyway, it was fast, and on fairly modest equipment: A DG Aviion with first generation Clariion disks, 512 Meg of Ram and 4 CPU's.

This may be different for really large OLTP databases with a very high number of users, but I've never had the privilege of working on one that big. e.g. Amazon.com, etc.

Jared

>
> > If you're familiar with the Help Desk software 'Remedy', you will know
> > that it has one of the worst schemas ever designed by man or beast. If
> > you haven't seen it, you would have a hard time imagining it. Yes, worse
> > than Finanacials, Lawson, SAP, etc.
>
> My general impression is that big, commercial packages tend to have a
> really bad reputation for being implemented on bad models. If I stay
> here, I'll probably learn more about Peoplesoft over the next couple
> of years.
>
> Obviously it is human nature for multi-platform commercial package
> vendors to not "do the right thing" from a platform oriented purist
> dba perspective.
>
> But, I can't see that all that has anything to do with legitimate
> structured denormalization practices.
>
> What I'm mainly thinking of was an old article in Oracle magazine on
> denormalization by Ulka Rogers.
>
> > ( 'where is he going with this?' you ask )
> >
> > One of my fantasies is to build a help desk system that runs on a
> > normalized schema, open source it, and put Remedy out of business. The
> > schema is that bad.
>
> Blessings and peace be upon the righteous and those that abide by the
> will of the Supreme Being.
>
> > > pss, aren't you *ever* going to tell us what happened at your last
> > > job?
> >
> > Sorry, thought I had.
>
> sorry, I must have missed it. I try to always read your posts, but
> sometime, I just have to hit the filter key for this list, and I
> never have time to go back and read everything that is dumped out to
> an archive folder.
>
> > My previous employer laid off several folks. I wasn't among them
> > however. Damagement decided to take this opportunity to redeploy several
> > positions to HQ in Houston TX.
>
> hummm, "energy crisis" mess? :)
>
> > If you've spent any time in the Pacific NorthWest, you may understand why
> > I chose to stay here. Likewise if you've been to Houston :)
>
> Yes, I have relatives that were in Eugene from the late 1970s until a
> couple of years ago. Excellent quality of life if compared to the
> "rat race" mentality prevailing in most urban centers. But, it is too
> darn drizzly up there and claustraphobic with all those big deep dark
> rain forests for me personally, I'd rather be in the Sonoran desert
> (eg, Tuscon) if I had my 'druthers.
>
> > ( hope I didn't offend any Texicans :)
>
> technically I'm a 5th generation (unoffended) texan, but my parents
> got me out of there prior to my 2nd birthday. I do like grits,
> bisquits and gravy, blackeye peas, okra and all that good stuff. :)
>
> > I'm taking this opportunity to attempt a slight career change and get
> > into the contracting side of things.
>
> wow!
>
> thanks/regards,
> ep

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Jared Still
  INET: jkstill_at_cybcon.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).