Something so timecritical should not be subject to the...

Put a PRINT statement prior to the EXEC.

PRINT @Query

EXECUTE sp_executesql @Query

Run the batch. Copy the output from the Messages window and paste it here.

If I do this, you will remain a novice. You must at least make an attempt to write the TSQL.

Please post ddl and dml for the tables involved in your query.

This is quite a trivial request and there are several ways to do it - what have you...

Pass the string into the stored procedure as a string parameter and split it there. Splitting it first and attempting to pass in all of the resolved elements as a...

That's okay, it was always going to be a long shot with no sample data blah blah.

I hope you can take away a few ideas from it though, and...

Heh I see it now.

"In the...

You might get some performance lift by jigging around with your filters. If you're going down this route then do take on board Gail's comments about DISTINCT. You might benefit...

Always carefully test any solutions offered to you. The one you rejected works with the sample set, the one you appear to have accepted does not. Here's a third offering:

SELECT...

-- The requirement is the Table1.SetId (1, 2, 3...) is to be assigned to the Table2.SetId

-- for the same number of rows as the value of Table1.MsgCount. For example,

CREATE...

SELECT

s.subTypeId, s.subTypeValue, s.subparentId, s.childLevel, s.sortOrder,

s.leafFlag, s.aggregatesSubTypeId, s.sourceIdentifier, s.origId, s.RecId,

camasapavalue = '[' + x.ConcatenatedString + ']'

FROM #SAMPLEDATA s

CROSS APPLY (

SELECT ConcatenatedString = STUFF(

(SELECT ',' + CAST(subTypeID AS VARCHAR(30))

FROM...

Can you post up some sample data, Kapil? Showing multiple rows per key? Obfuscate if necessary, but it really would be very useful to see it.

Let's say you have a "dupe pair" of rows in your source table - dupes on the basis of a unique key on one or more columns.

Does it matter...

MERGE #temp T

Is the table #temp always empty before you start the upsert?

Use a traditional upsert.

What you're trying to do in your original MERGE is to update rows which don't exist.

Perform your inserts first from a deduped source set -...