Sean Grebey (7/2/2013)
I was asking because from what I understand if there is a clustered index non-clustered indexes use the clustered key, otherwise it uses the rowID (file number, page number and slot number), so it would seem that having a clustered index would improve performance of the non-clustered indexes but I haven't actually found anywhere that said so.
It's not possible to make that declaration without more information, e.g. the number of columns in the clustering key, its data types and whether it is declared as unique or whether it is actually unique, the data access patterns of the table. Focus on picking a good clustering key, one that is narrow, unique, preferably ever-increasing and most importantly one that supports your most common data access patterns.
There are no special teachers of virtue, because virtue is taught by the whole community.
--Plato