Have you checked out c2 auditing?

Assuming that the join is sargable then yes (not converting column data in the join).  The optimiser is smart enough to figure the best path.

... and I would normally agree with Charles (if only 300 k /day) then you don't need more than one table / month but this is a monster database.

That's why I proposed the cude idea... (assuming it's possible)... solves the multiple queries part and the long wait time.

If I remember correctly, it's 10-15 M/day. (Phone company).

And folks, that's why you should always post the code you put on the forums.

But I'm sure you can figure out why this doesn't work by yourself now.

Doh, operations priority... the cast is made after the division; easy fix :

SELECT 1.0 * COUNT(cdu_slaok) AS Total , SUM(CASE WHEN cdu_slaok = '1' THEN 1...

You can't do that on sql server like Acces let you do.  You must reuse the count and sum operations for the total >>

SELECT COUNT(cdu_slaok) AS Total ,...

Oops, ya works better like that

SUM(CASE WHEN A = '1'

Well you could wrap an if around the dbcc command >>> 'Use ?; IF ''?'' not in ('tempdb', 'Pubs', 'NorthWind') begin DBCC CHECKDB END

You can pass pretty...

... and also if this is a production server, I don't see why Pubs and Northwind are installed .

Did you try my query??? It's what it's supposed to do!

Now with testing

EXEC

dbo.SP_MSForEachDB 'USE ?; DBCC CHECKDB'

So you have a boolean char .

SELECT COUNT(A) AS Total , SUM(CASE WHEN 'A' = '1' THEN...

Something like this might work (untested)

EXEC

dbo.SP_MSForEachDB 'DBCC CHECKDB'