Re: analyze index

From: Igor Racic <igor.racicC_at_freesurf.fr_no_C>
Date: Thu, 03 Feb 2005 21:09:48 +0000
Message-ID: <42028408$0$27049$626a14ce@news.free.fr>

  Hi, Jonathan, hi, Richard,

  I don't know how to get that situation - probably it's a bad day :-)   Do you have some simple scenario ?

  Thanks
  Igor

Richard Foote wrote:

> "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote in message 
> news:ctee78$d0s$1_at_titan.btinternet.com...
> 

>>Note in-line.
>>
>>--
>>Regards
>>
>>Jonathan Lewis
>>
>>http://www.jlcomp.demon.co.uk/faq/ind_faq.html
>>The Co-operative Oracle Users' FAQ
>>
>>http://www.jlcomp.demon.co.uk/seminar.html
>>Public Appearances - schedule updated Jan 21st 2005
>>
>>
>>
>>
>>"Richard Foote" <richard.foote_at_bigpond.nospam.com> wrote in message
>>news:A9LJd.135384$K7.133006_at_news-server.bigpond.net.au...
>>
>>>- In reality, such a pattern as you've described is unlikely and would
>>>require a rather manufactured insertion method by the application. That
>>>said, factors such as specific deletion/insertion patterns, the possible
>>>effects of multiple freelists and the possible effects of ASSM could
>>>possibly lead to similar CF patterns.
>>>
>>
>>On the contrary - set freelists = 2 on the table, then insert rows which
>>are
>>keyed on (trunc(sysdate), sequence_number), and you get that sort of
>>pattern almost automatically.
>>

> 
> 
> Hi Jonathan,
> 
> Your right.
> 
> I should have said something like with no freelists, blah blah is unlikely. 
> That said, factors such as freelists, blah blah could possibly lead to 
> similar CF patterns (which was my rather clumsy attempt).
> 
> Thanks for the clarification.
> 
> Cheers
> 
> Richard 
> 
> 

Received on Thu Feb 03 2005 - 15:09:48 CST