I think you've got the computed column backwards here.

That is, since the data is supposed to be in 1NF, the "divided by 10" value should be in a...

Some refinement is possible.  For example, for any single column that has a unique index on it, you wouldn't need to count the values, instead simply get the number of...

One of the main reasons to not give VIEW DEFINITION to users is so that they can't see the underlying physical table names.  If you let them know the physical...

I'd say you get a slight gain by disabling the non-clus indexes first.  That's because disabling the index requires removing the rows, which requires some minor logging (of page deallocations,...

Or just:


SELECT LEFT(OrigDateString, 4) + ' Q' + CAST((CAST(SUBSTRING(OrigDateString, 6, 2) AS tinyint) + 2) / 3 AS varchar(1))
FROM #TestData

Columns lengths have another effect on memory allocation as well.  SQL reserves memory prior to processing a table.  It determines the amount of memory to reserve partly based on column...

Or:

select *
from lineitems
where orderid in (select orderid from orders where customerid = @CustomerID)

I think it's very possible that o.type = 'U' would speed it up, rather than using OBJECTPROPERTY.

Move the error details to a different table.  That table could be keyed by the clustering key of this table plus perhaps a sequential error number.

This table should...

There's a fundamental difference between those queries.
The first is an INNER join to dept, obviously, as explicitly stated, whereas the second is a LEFT OUTER JOIN, although only implicitly.