Re: Thee is a problem here I am sorry to have to report.

"Jitendra Patel" <coldrainandsnow_at_lycos.com> wrote in message news:7610580f.0310120446.27d63a93_at_posting.google.com...
> > Needless to say there is an obvious problem here. One that goes far
beyond the previous concerns about trolling, insults, and abusive off-topic remarks.
>
> Mr. Morgan,
>
> Observe below is it is FORUM MEMBERS doing the attacking.
>

Hi Jitendra,

You really have no idea do you ?

How about a little technical challenge !!

How about you defend the TUSC God you worship from these awful "attacks" (your term) you so denounce by suppling a technically factual explanation/defence instead of the boring and rather juvenile tripe you usually dish up. Here's a chance to prove the TUSC publications correct, that you know something vaguely about Oracle, actually post something technical and earn some respect, something you're sadly lacking at the moment.

Two very simple challenges. Explain why the following claims printed in TUSC 8i performance tuning books are correct and undeserving of these awful recent "attacks" (your term):

Challenge 1) "The binary height of an index increases due to the size of a table and *the fact that the range of values in the indexed columns is very narrow*".

Please explain how the range of values in an index influences the height of an index ?

Challenge 2) "Generally speaking, fragmentation decreases performance by 10-20 percent on average. To repair fragmentation, you'll need to find the fragmented tables/indexes. It's recommended that you regularly monitor your database to find tables/indexes fragmented into more than five pieces (or extents). The key to eliminating fragmentation is rebuilding the table to one extent or the size it has currently grown to."

Please explain how having five extents decreases performance by 20% and how having just one extent subsequently improves things by so much ? Remember this is in a tuning book that covers 8i but to make things easier for you, please justify such claims with either dictionary or locally managed tablespaces. You don't even have to concern yourself with the tablespace fragmentation such steps would generate in DMTs.

Here's your chance at last. To post something that's actually related to Oracle, that supports TUSC teachings and that supports the above claims as being correct.

Because if you can't, if you agree that the above is crap, then surely you support these "attacks" (your term) as being justified and that exposing them as technical illusions is in the best interests of the Oracle community and that such "attacks" (your term) may not actually be levelled directly at TUSC but at the information itself and at the confusion and misunderstandings that such inaccuracies generates.

So Jitendra, are you up to the challenge ?

Cheers

Richard Received on Mon Oct 13 2003 - 08:28:13 CDT