Luis Carlos Junges wrote:

> thanks for the explanation.. now... i looked at the indexes and found that the biggest one that must be reduced and, consequently, will make the procedure quicker.
>
> TABLE:tbrelshape
> BRUSHID
> SHAPE
> WIDTH
> LENGTH
> THICKRADIUS
>
> Primary key(brushid,shape)
>
>
> the index has a selectivity of 0.083333 and the others 20 indexes used at the procedure have value lower than 0.005...
>
> the index is the shape field from the table which is a foreign key to a table TBSHAPE will all possible shapes.
>
> the population of the table is as follow (brushid is always different):
>
> 1 = CON
> 145 = CYL
> 551 = PAIR
> 78373 = SINGLE
> 20207 = SPLIT
> 16 = TRIPLE
> 256 = TSPLIT
> 1 = WEDGE
>
>
>
> well, running the procedure for shape TRIPLE, for example, it is really quick while for SINGLE, the problem i am talking about appears....
>
> i will thanks for any idea of how i should modify the indexes in such order to get a procedure always fast as possible Or how i could organize this little data in such way to get a procedure always fast...
>
>
>
>
>
>
>
> ---
> Luís Carlos Dill Junges ©
>
>

Wich query ?

Wich execution Plan ?

see you !

--
Alexandre Benson Smith
Development
THOR Software e Comercial Ltda
Santo Andre - Sao Paulo - Brazil
www.thorsoftware.com.br