The adhoc queries will cost more in terms of compile time but that is probably small compared to the read time. They may be getting autoparameterized anyway if they...

You may be fine, but it depends on the workload.

Because SQL is reading an extent at a time from disk, you wouldn't want the average fragment size to...

The first thing I noticed is that you have min memory set to 200000MB .... a bit larger than you would like! That shouldn't have any bearing on your...

Ha, as you can see I edited my post as you were typing! Of course the line I missed originally changed the complexion of the question ...

Anyway to clarify...

Bruce: thanks for the quick reply.

It's not exactly what I expected, in that by using FOR EACH ROW it's no longer a set-based trigger ie one that can...

Have any links / examples? This is perfect timing for something I'm working on.

I added the -h parameter to sqlcmd to drop headers. You also need to set nocount or parse out the returned values to only return valid instances.

function...

I haven't tried it for your specific code, but piping to out-null should do the job.

gcopeland (1/12/2009)

Absolutely .... however you...

I can't believe I read this entire thread.

It seems to me that everyone (some more than others) will judge by appearance at times. What matters is whether...

I think the point is that if it is known that there will be a significant number of records updated with no change, why update them in the first place?...

It can be done, but you would need to keep an account on each machine with the same id (and keep the password synchronised). Alternatively could you run the...

Glad you got it working .... I don't think it had much to do with me though 🙂

Having seen your post that lead to this one, is your real problem with the performance or in the space taken in tempdb? If it is just space taken...

Are the columns in the two indexes in the same order? That can also impact the join type.