Re: Theoretical definition for the number of unique values?

On Apr 27, 9:01 pm, Lemming <thiswillbou..._at_bumblbee.demon.co.uk> wrote:
> On 27 Apr 2007 17:52:00 -0700, hpuxrac <johnbhur..._at_sbcglobal.net>
> wrote:
>
>
>
>
>
> >On Apr 27, 8:21 pm, Lemming <thiswillbou..._at_bumblbee.demon.co.uk>
> >wrote:
> >> On 12 Apr 2007 18:23:07 -0700, sqlservernew..._at_yahoo.com wrote:
>
> >> >I found out. It is called "COLUMN CARDINALITY"
>
> >> >Sorry, no prizes.
>
> >> >http://www.informatik.uni-trier.de/~ley/db/journals/tods/WhangVT90.html
>
> >> >(1) obtaining the column cardinality (the number of unique values in a
> >> >column of a relation) and
> >> >(2) obtaining the join selectivity (the number of unique values in the
> >> >join column resulting from an unconditional join divided by the number
> >> >of unique join column values in the relation to Be joined).
>
> >> >These two parameters are important statistics that are used in
> >> >relational query optimization and physical database design.
>
> >> >http://www.idig.za.net/mysqlindexes/2006/11/09/
>
> >> >Column cardinality. This is the number of unique values contained in a
> >> >column. Indexes work best when there is a high cardinality. Put
> >> >another way, the more unique values there are (fewer duplicates) the
> >> >better that column will be for indexing. Consider the ID number column
> >> >of the previous example. Here there are no duplicates, only unique
> >> >values. This column will be ideal for indexing. On the other end of
> >> >the scale may be the first names column. Here there will probably be a
> >> >number of duplicate names (fewer unique values) and a lower
> >> >cardinality compared to the ID column.
>
> >> Yes, cardinality is the correct term.
>
> >> Now, for bonus credits: can anyone tell me the correct term for
> >> someone who posts a homework question here, gets an answer, and then
> >> pretends he worked the answer out for himself?
>
> >> Lemming
> >> --
> >> Curiosity *may* have killed Schrodinger's cat
>
> >Way to jump all over a thread that died 2 weeks ago.
>
> Mate, most of usenet died more than 2 years ago. What does it matter
> if I'm reviving someone's fortnight-old homework? Especially if I am
> taking the piss.
>
> Do try to keep up.
>
> Unless, of course, it was *your* homework? Forgive me if so; I can't
> be bothered to read back. But I can understand why you might be
> feeling a bit sensitive about it.

Not exactly.

Different people pick different tools to read these posting.

The cdos group is still very active. This item was cross posted to various groups and was effectively dead until you chimed in.

Personally I use the google groups interface. If you take a look at that tool you might have a different opinion about the health of what used to be usenet. Plus it allows you to see the question from the op, the replies and the thread in context.

Many of the other people responding in cdos use other tools. Received on Sat Apr 28 2007 - 13:07:34 CDT