Re: Where is Oracle’s Grid ?

"Daniel Morgan" <damorgan_at_x.washington.edu> wrote in message news:1071847804.695870_at_yasure...

> The seem to be blaming the wrong party. Oracle has provided the tools to
> the 3rd party vendors. Do you expect Larry to send over a couple of
> heavies to force them to use the tools properly?

They are blaming nobody. The v$stuff is just essentialy useless to a production dba dealing with one of those 3rd party monsters, that's all. Other than to tell them what they already know: the app software is crap.

> >>Let me guess: Siebel? SAP? PeopleSoft?
> >
> > All of them.
>
> <colorful comments and gestures made here>

Absolutely. The only one that is remotely acceptable is Peoplesoft, it has some good points. The other two are just ports of glorified mainframe flat-file shite, 30 years old in design. It's a rip-off of monumental proportions!

> I still fail to see why it is Oracle's fault that people used their tool
> set to build junk? Or perhaps more sharply ... why it is Oracle's fault
> that your organization didn't properly test the product before laying
> out such huge sums of money.

You never saw an Oracle sales rep recommending someone buy 3rd party crap just so they can get the RDBMS licence? Seen quite a few in my time...

> I'd buy your argument if it was impossible for anyone to build a
> third-party application using the Oracle database but it isn't. And I
> know for a fact there are a huge number of extremely successful projects
> built with the same starting tools used by Siebel, SAP, and PeopleSoft.

What's that got to do with the tuning tools provided by Oracle being completely useless to users of these 3rd party shite? The problem is NOT that Oracle doesn't have the toolset to make good apps. People making Oracle apps out there are however the vast minority of users.

The tuning tools put out by Oracle over the years have always assumed that source code for the app is available. To be changed, to be tuned. That hasn't been the case since the mid 90s! With dire consequences to DBAs that have to deal with this 3rd party rubbish.

The current tuning tools from Oracle are essentially useless other than to confirm what everybody already knows: that the apps are extremely badly designed. In what way does confirming this help the poor sod who has to put up with their inneficiency?

> > Of course not. When I buy Oracle, I don't expect them
> > to come over and run things for me. But that's what they
> > want to do now.
>
> In what way?

By asking the customers to give up their own data centre and DBA and offering them the Oracle service bureau instead.

> > Of course not. But Oracle wouldn't mind keeping the steak
> > in-house...
>
> But doesn't.

Of course it does. If they could, they'd ask every customer to outsource their IT to Oracle.

> > I don't expect them to teach me anything! Most DBAs out there
> > don't NEED instruction on how to run the product. Like anyone
> > else, they are perfectly capable of deriving it from a manual,
> > or a book, or a white paper.
>
> Agreed. But I thought that was your complaint.

No, not at all.

> That somehow it was
> Oracle's responsibility that 3rd party app vendors didn't use the tool
> properly.

No. It is however Oracle's responsibility to provide DBA's with tools to tune crap SQL without having to have access to the source code.

Which Oracle has NEVER done until 10g. And that one is open to verification. Everything else they have done in this subject assumes that there is SQL source and other source to be fixed.

Which is simply not the case for the vast majority of users out there with 3rd party apps. And hasn't been the case for the last 8 years!

So, do you still think DBAs out there who have to cope with this situation are the ones that don't use the "features"? Here is something then: next time someone buys SAP against your recomendation, instead of throwing in the towel and leaving, stick around and watch what happens when that crap gets some data load.

Then when all hell breaks lose and management starts to look for scapegoats (because management is NEVER wrong, it's ALWAYS someone else's fault!), you step in and run your V$stuff. See how far you get in tuning something that takes 3 days to run by moving it to a different buffer! Without a single chance of changing the source or modifying the design. Then come back here and tell me the tuning "tools" from Oracle are good!

> > What they cannot do is go against a 3rd party vendor without
> > Oracle's support. And that support has over the last 5
> > years not been there at all. Since Oracle decided to get into
> > the service bureau market.
>
> I'm lost with the above comment.

Oracle has been asking their customers to outsource their IT involving the use of their RDBMS to Oracle's own consultancy/service bureau. What, you didn't know?

> That may be the flavour in Oz but certainly not here. Here you will more
> likely find DBAs having little or no influence over the database that is
> purchased. They are implementers not decision makers.

Same here. But they get asked for an opinion. And until a few years ago they'd be the first to try and convince their bosses to stay with or go with Oracle. I haven't seen a single one do so, in the last 3 years...

> And from the
> numbers I am seeing Oracle is actually increasing not decreasing its
> market presence.

Wait a little more. This is not micro-economics.

> Is it the policies survival ... or the ability to change those policies
> in accordance with business conditions that matters most? Survival
> usually goes to the niblest.

When you're talking strategic buying, it's policy survival that counts. No one selects db makers for 6 months anymore. Survival policies deal with that sort of timeframe and are useless nowadays. We're back to the old pre-IT boom days now. Remember? You should, both you and I were around then.

-- 
Cheers
Nuno Souto
wizofoz2k_at_yahoo.com.au.nospam