Hello Grant Fritchey,

thank's a lot for your very detailled explanation. I'm still not sure if I understand the part I'm interested in most.

I know these queries had no predicate and I'm aware of the differences of index scans and index seeks. The problem I personally have is that many times I see index scan operations and many times I'm not sure if it scans the whole index or just a part of it.

If I use the following query:

select top 1000 * from #TestIndexAccess

I still have a clustered index scan but it is NOT an ordered scan. Still the first query needs only 3 logical reads. I don't think it did a full scan. Now how do you see the difference of this last query to the others?

If I have a table with several million of rows and the partial scan returns some 100 thousands of them I dont's see a possibility to determine which kind of index access happend.

Now, did I not understand all of your information or did I forget something else? 🙂

Thank's a lot, Wolf