Are you sure about that? 😉

Good news. I hope all goes well.

Just for giggles:

-- Source table

CREATE TABLE #Table1

(

row_id INTEGER...

SQL-Student (3/15/2010)

---

What can be done to prevent the deterioration in performance when the full backup runs? Since the backup process itself will generate lot of IO to slow down...

One more way:

SELECT -- Use a column list

*

FROM (

SELECT ...

[font="Courier New"]Msg 207, Level 16, State 1, Line 6

Invalid column name 'ttl_av'.[/font]

There are 'better' ways, but this will work:

SELECT ib.account,

...

In your second post, yes. That's why I asked if there was...

You are right of course. I have updated my code sample to avoid confusion. Thank you.

Roust_m (3/14/2010)

---

I am just wondering if there is a way to find out the amount of memory a query needs (estimated and/or real) and the time the optimizer thinks it...

Simple demo:

DECLARE @demo

TABLE (

group_id INTEGER NOT NULL,

value ...

No need for a loop.

-- Test table

CREATE TABLE #Sample

(

column1 ...

OK, all done. I had to include the RAND in the wrapper view too, for the same reason.

Here is the full script:

USE tempdb;

-- Drop the in-line function

IF ...

There is another one inside the CROSS APPLY. Not difficult either, I am just hoping the query plan holds together.

The article would have benefited from code examples, either in-line or in an attached file.

The only type of denormalization considered is full redundancy: storing the results of a complex join...