Re: Seeking Advice: MS SQL DBA learning Oracle?

Well, I didn't mean to limit it to developers. DBA';s should read the application developers guides also. It would be nice if managers had a clue, but often they don't. By performance I mean that if you know x is better than y then do x not y unless there is a really good reason not to do x. (eg "We'll put in bind variables later, when we get to the performance phase of the project." or "We will fix that problem where we read from the same table 120 times during a calculation when we get to the performance phase of the project." These are real comments.) Sure if you don't know then I can understand, but if you know you should do something basic in advance then do it don't put it off!

Jim

"Keld Nielsen" <keldnielsen_at_image.dk> wrote in message news:ORUD8.2311$4f4.170129_at_news000.worldonline.dk...
> Well, well, I have a DBA running around saying '*don't use* nvl and
decode.
>
> Can you imagine why ? (comment was: don't use them in a where clause and
> by the way what's a function (further on - what's the difference between a
> function
> and a procedure - no, I'm not joking !))
>
> I do somewhat agree with your comments below Jim, but I disagree that
they
> only
> apply to developers.
>
> When has it ever been possible to include sufficient time
> to 'put performance in from the start' ?
> (the answer I always get is 'cut your estimate by x%')
>
> Keld.
>
> "Jim Kennedy" <kennedy-family_at_attbi.com> wrote in message
> news:38QD8.41666$WR1.39058_at_sccrnsc01...
> > Dusan,
> > Right on. Yes, the developer should have at least a basic understanding
> of
> > how the backend works. Just the other day I had a developer want to
> create
> > temp tables because that's what you do in Sybase. I spent some time
> > explaining Oracle to them and they were pleased as punch that they
didn't
> > have to deal with that mess anymore. What also surprises me is those
> > developers that don't read the APPLICATION developers guide. So they
don't
> > put performance in from the start - "we will tune later." Performance
is
> > part of the daily job not something you bolt on later. Okay, I
> occasionally
> > use the _run_faster=true parameter, but usually I make the developers
fix
> > it. :-)
> >
> > Jim
> >
> > "Dusan Bolek" <pagesflames_at_usa.net> wrote in message
> > news:1e8276d6.0205130013.234fb31a_at_posting.google.com...
> > > rgaffuri_at_cox.net (Ryan Gaffuri) wrote in message
> > news:<1efdad5b.0205121459.cd19ffe_at_posting.google.com>...
> > > > You have IT experience which is VERY good. However, you must start
as
> > > > a developer and NOT as a DBA. There are no entry level DBA jobs. You
> > > > have to be more of a programmer first for 2-3 years then you may be
> > > > able to move over. Anyone who says you can start as a DBA is either
> > > > lying or an idiot.
> > >
> > <snip for size>
> > > Yes, I'm going to flame you right now !!! ;-)
> > > For all developers: Please read Oracle Concepts. You should know
> > > something about architecture of database you're writing for. I'm tired
> > > about all that worthless talks about purpose of schemas, why database
> > > != schema, why more than one schemas can be in one tablespace, why is
> > > not a good idea to put two 30gig database in one instance even if
> > > they're on the same DB server etc ...
> > > And performance tuning is indeed the must for any Oracle developer.
> > > New developer should start with learning about performance impacts of
> > > various SQLs as soon as wrote the first lines of code. Oracle
> > > developer without tuning skills is worthless and even dangerous.
> > >
> > > --
> > > _________________________________________
> > >
> > > Dusan Bolek, Ing.
> > > Oracle team leader
> > >
> > > Note: pagesflames_at_usa.net has been cancelled due to changes (maybe we
> > > can call it an overture to bankruptcy) on that server. I'm still using
> > > this email to prevent SPAM. Maybe one day I will change it and have a
> > > proper mail even for news, but right now I can be reached by this
> > > email.
> >
> >
>
>
Received on Mon May 13 2002 - 20:50:44 CDT