It's not the same query if you've changed the code.

However, yes, with the same code, same data, same statistics distribution (you won't get the same statistics) you could get different...

Is it a table valued UDF or inline scalar? If it's table valued, that might the core issue righ there. And no, that's not unique to 2005.

Even if it's inline,...

If you want to get some idea as to why it works differently between 2000 & 2005, get the actual execution plan and see what SQL Server is doing with...

It's relational. I don't know enough to say it was poorly designed, but eliminating FK's...

If you're running a query through a linked server, most of the time, all the data is moved across the wire first, then filters & joins are done to extract...

I'd have to say it falls into it depends. I don't mind reading a bit, but sometimes people post several hundred lines of code and, quite frankly, I don't want...

That sure sounds like a pretty standard INNER JOIN.

Try this:

CREATE TABLE #Parent

(ParentId INT IDENTITY(1,1) NOT NULL

,Val NVARCHAR(50) NOT NULL

,CONSTRAINT pkParent PRIMARY KEY CLUSTERED (ParentId))

CREATE TABLE #Child

(ChildId INT IDENTITY(1,1) NOT NULL

,ChildVal...

DISTINCT is killing your peformance. You should try to find away around it. Plus, from the execution plan provided, it was five rows going in and five rows coming out....

If your page life expectancy is high, there's not to do but sip some coffee and look for other causes of poor performance. A high page life means that the...

Someone or some process such as a SQL Agent job must be switching it to simple recovery. I'm not aware of anything that does it automatically in the background.

Which model...

Darn. And I'm hungry too. Ah well. Life is full of these little disappointments.

First round of scotch is on me!

I agree with the others but feel the need to express it a different way.

If you're simply learning SQL Server, no. It's conceptually beyond the basics a bit.

If you're using...

I think my issue with doing it the "interesting" way as opposed to using JOINs is that, while this time, the performance and plan is likely to be identical, far...

Jeff Moden (4/11/2009)

---

Well... I did it. Thank goodness they extended the deadline of PASS 2009 Session Abstracts. I'll be there one way or another this year.

I...

radek (4/11/2009)

---

Hi,

really you need all columns (*)

So you can create nonclustered index on macaddress is key of index and this index will have inlcude other colums, try this....