Can you create an Archive database (or table), move the data there, delete it from the main database, and then the reports that might need the old data can use...

I fixed the second insert statement (needs a null or ID for the third column in the first select), ran the code, and got results that seem to make sense...

Ron Kunce (7/7/2008)

---

I would like to know if there could be any possible advantage to having an identity primary key field set as non-clustered and having a separate clustered index...

The ANSI_Padding option would seem to allow trailing spaces to be handled the way you want them, but I just tested and it doesn't do it. It considers the...

I read your question about firing for each row differently than Steve did.

Triggers will fire for ALL rows, but not for EACH row, if you get the difference.

If you use...

You can do this with a simple set of left outer joins, I would think. Otherwise, it's Union All statements.

To be more specific, I would need to know the...

Generally, bulk insert doesn't fire triggers. But it can, if you set it to do so. For example, the Bulk Insert command has an option "With Fire_Triggers". ...

If you can't replace the field with nvarchar(max), instead of ntext, then you'll probably need to use the TextPointer and WriteText methods. Look up those keywords in Books Online,...

What columns would those be in other than the ones you don't want?

Run a trace, find out what's taking that long.

I've been in environments where security was severe, where security was lax, and where security was misapplied.

In the severe security environment, everyone from the bottom up paid attention to it....

Just for the sake of curiosity about performance on this whole thing, can I ask you to try out this:

;WITH

TopBene (BenefitType, BenefitID) as -- Selects top ID by type

(select...

Brian Tax (7/8/2008)

---

The plan changed slightly, for the PersonalData Index Scan it is now using the new index instead of a dta_ index. It is still not doing a...

I've had a similar problem, but it solves itself if I click on the instance to open its object tree. I did pretty much the same thing as you,...

The specific columns plan has an index scan of PersonalData, while the * plan has a RID for that table based on an index seek.

I'd see about doing a covering...