Peso, your returns an error.

Where does this 'golden rule' come from? What does it give you?

During the life of these two tables this could change.

What will happen then?

Have a go at this. It will even find duplicate maximum concurrent users. Don't know how this will run with millions of records.

select s.Date,s.Application,m.maxusers

from

(

select convert(varchar(10),Date,120) d,Application,max(Concurrentusers)...

You need to do a group by convert(varchar(10),Date,120),Application

Ran some simple tests (A=25K records B=780K records) and found no difference in performance in the two variants of the sql syntax that use INNER JOIN.

Lookup sp_executesql in the BOL.

For example:

declare @s-2 nvarchar(4000)

set @s-2='select * from sysobjects'

exec sp_executesql @s-2

Examples 1 and 2 potentially produce two different results.

Same for 3 and 4.

Regarding 5 and 6, the result is always the same. So do you want to know if performance...

So now I know the speed of your computer - give or take a milisecond. Pretty soon I'll know what color it is.;)

I finally understand your 'wicked' update in Peso3....

And Peso1? Or are you still waiting for it to come back?:D

What do you mean by 'making blog'. Where and how?

Peso, taking a hard look at your query and wondering why you are linking on ProcessCell I realized that I had misunderstood the op's problem. I didn't understand that each...

I hope you were able to use the SS 2005 solution.

When limited to SS 2000, our queries really start to slow down after a while.

Jeff, this generates random intervals based on some parameters. Give it a try.

Peso,

Aaahh, I didn't know you are a performance freak. The last time I did some performance tests was on a thread dealing with hours between two dates exlcuding weekends. Jeff's...

Peso, when I run this no rows are generated.

I tried it on SQL Server 2005 and Oracle 10g (10.2.0.1) and I had to change nothing with respect to what I described previously.

I'm not too clear what you mean...