Well, now that you talk about it, maybe you can give me some more input. (I saw you are currently running another thread about this).

In the books on-line, I saw the pros and cons of having a clustered index.

In my case I have some 4000 tables. So theres no point on creating, now a clustered index on each of them.

All my tables have a unique index on a stamp column that I generate acording to an algorithm of my own. This key is 30 characters long. As I read, creating a clustered index on this column is not as eficient as it could be because they are all discrete values.

Furthermore, creating this clustered index, would make SQL Server use this keys in all other indexes (to point to the real data). Being such a long key, this is not good.

These are the reasons that led me not to use a clustered index on theses columns.

The other indexes I have, I create and maintain them all with the same functions. So (as I can only have one clustered index per table) thay aren't clustered indexes.

Although I might use a clustered index for specific cases in the future, I decided not to use them in general.

What do you think ?


Thank you in advance Mike.

>>> I analysed the case and decided that it was not positive, in my case, to have a clustered index in each table.
>
>Why? The clustered index is the most powerful index that SQL Server supports.
>
>-Mike