Re: Index management

Don !!

Welcome back, we've all missed you, we really have.

I'm really looking forward to reading your post, it's sure to be full of accurate, interesting, technical information ...

"Don Burleson" <don_at_burleson.cc> wrote in message news:998d28f7.0405071236.7a48987a_at_posting.google.com...
> > In the context of the present discussion, however, was is most
> > interesting is the distance that apparently now exists between Don and
> > Mike on the issue. I hope the divorce won't be too messy.
>
> As usual, totally wrong. . .
>
> Actually, anybody with the balls to bother arguing with you losers
> should be commended.

Now Don, that's not very nice (and not very professional either).

We're not arguing with Mike, we're just trying to show him why he is so utterly wrong. Hopefully, it might eventually sink in.

Why are you calling all of us "losers" ? It seems a remarkably ungracious thing to say considering all the many times your "expert" articles have been discussed here and considering the many times we have highlighted the numerous errors that they've often contained. These highlighted errors have been most beneficial to you because of the numerous times they have led you to correct and in some cases totally rewrite your articles as a consequence.

Don, please tell us. How many times have you *directly* made amendments to your articles based on what I and others have discussed here. Please Don, how many times ? And how many of those times have you publicly acknowledged these contributions ? Please Don, how many times have you thanked and publicly acknowledged changes you're made to your articles based on discussions here ?

I really don't think "losers" is a nice thing to call people who have helped with your knowledge of Oracle do you ?

>
> In fact, I'm so impressed with his posts here that I just hired him.
> He starts in 3-weeks:
>
> http://dba-oracle.com/oracle_news/2004_5_11_burleson.htm
>

So you've hired Mike to work for you based on his posts here in the last few days ? That's quite remarkable !! Not sure how many other people would hire someone who :

• has no idea how Oracle indexes reuse space and thinks index entries are only reused if identical (as per his quote he's attempted to defend)
• incorrectly thinks that the CF is affected by index splits
• incorrectly thinks that the CF is affected by how "broad" an index is
• incorrectly thinks that the CF is affected by DML changes to the *indexes*
• incorrectly thinks that an index should be rebuilt based a ration between the CF and the blocks in a table
• incorrectly thinks that an index rebuild will improve or alter the CP
• in summary, has no idea about the CF
• thinks that indexes should be rebuilt based on x (2) levels in an index (also per the quote he's attempted to defend)

What's perhaps worse, is that when faced with the cold hard facts of logic and the technical truth, he either chooses not to or is incapable of accepting it.

And this has *impressed* you ? Only you Don, only you. You have an interesting selection criteria to say the least ...

> If you want to retain him to explain index internals to you, here are
> his rates:
>
> http://www.dba-oracle.com/consulting_prices.htm#ault

You mean you're going to charge people to have Oracle myths in relation to index internals explained to them !! That doesn't sound quite fair to me ...

I tell you what Don, here's the deal. I'm going to give you the perfect opportunity to prove all your doubters wrong. This will be the perfect opportunity to show to the world what an Oracle "expert" you really are. You too make the claim that the CF is a reason for an index rebuild, your memorable "Index Internals" article said so as per your quote in my presentation (what do you think of my presentation btw).

All you have to do is simply explain to us all how Mike is right and I'm wrong.

1. Why, why should the CF improve when you rebuild an index ? What are the technical reasons for such a phenomenon ?
2. How, how does a rebuild that results in no change to the index order result in an improved CF ? What are the magical steps that Oracle performs, please a technical description ?
3. Where, where is a demo that proves that a rebuild changes the CF ? Where is your proof ?
4. and finally in relation to your specific quote in my presentation, please explain how an Oracle index can become unbalanced. Please explain to us all why and how an Oracle index will "spawn to a fourth level only in areas of the index where a massive insert occurred, such that 99% of the index has three levels, but the index is reported as having four levels". How is this possible in Oracle ?

This is your chance to make a *positive* contribution by explaining why you and your new employee believe in such things.

And a final little question for you Don. How come most of all your previous posts here in these newsgroups have totally disappeared from the google archives ? Most of them have simply vanished, including your "classic" comments in the Index Rebuilding thread where you initially made your quote above. Where have they all gone ? Are you devastated that all your wonderful contributions have gone, or have you something to hide ?

Just interested Don.

But please, answer the questions above, I would dearly love to hear your technical views on these things ...

Cheers

Richard Received on Mon May 10 2004 - 09:28:10 CDT