Well if tempdb has grown to 29GB then it's a surefire signal that you shouldn't be running it with the default settings. Something at some point has needed that...

Is that the actual size of the data files or the log files?

I've seen interesting things with tempdb following runaway transactions whereby the log showed a negative available space value...

I guess it depends on whether you still maintain those backup sets etc. With log shipping in particular those tables can get bloated pretty quickly, so on my servers I...

Have you checked the contents of the table? If it already exists and has data in it, then you need to see if the duplicate values are already populated....

This should do it:-

SELECT STUFF('0630',3,0,':')

, CONVERT(time, STUFF('0630', 3,0,':'), 14)

In fact you don't even need to specify the 14 as the type of conversion; that's just me having a hangover from...

Ok, now that's definitely interesting, I'll try it on some larger datasets later with that sort of skew too and post back (just done this and it stays consistent with...

That's entirely my conclusion; however I've had people tell me that it shouldn't do that (unless they're using it in another way that means the filter isn't applied in that...

Though, personally, I'd really be considering if you need the procedure to be dynamic in that case, as it will now only work for any table that has the column...

I've not; but I think that's a little bit of my point - it isn't an even playing field, and despite APPLY seeming to be designed for this kind of...

Ok, so looking at what you've provided (and I'm making the assumption that you are only matching on StudentID in this case) If I run the following in debug mode...

Post the sample DDL statements to create the tables you are using along with some sample data. Then post the actual stored procedure code. As in the actual...

The elapsed time is per query, not cumulative. So the elapsed time is the ~1000 ms plus the ~660 ms; I'm guessing this is rounded up in the indicator...

If you create a MERGE statement that is going to delete a row, it will delete it....

If you don't want to delete the row at all and simply want to...

No worries; I think it's a case of finding out which rows contain the invalid character and then excluding them. Though I'd also be interested how they got in...

Try running SELECT USER_NAME(), SCHEMA_NAME() to see what user and schema your session is connected as.