Re: The Index Clustering Factor Explained

From: Noons <wizofoz2k_at_yahoo.com.au>
Date: Tue, 11 May 2004 21:38:22 +1000
Message-ID: <40a0bb81$0$21085$afc38c87@news.optusnet.com.au>

"Howard J. Rogers" <hjr_at_dizwell.com> wrote in message news:40a04df7$0$12393$afc38c87_at_news.optusnet.com.au...

>
> www.dizwell.com -> FAQ -> Performance Tuning.
>
> It *is* a bit of a rush-job, so I'm open to corrections and suggestions.

Is it possible to expand it for non-unique indexes?

The cf would be even more relevant there as it should tell us how many blocks may be visited when retrieving rows of a given value of the key. If the rows are adjacent on the blocks for a given value of the key, then it is a "good" cf. If the opposite (one row/block for all rows of a given key value) then the order of rows in the table may cause problems for retrieving all rows of a given key value.

That sort of thing.

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

Received on Tue May 11 2004 - 06:38:22 CDT