In general, I'd say this is a serious problem with the data or the structure, but, you could treat it like versioned data and simply follow one of the approaches...

I could be very wrong, but are you thinking of SQL Prompt from Redgate?

There are several ways to do this. I have an older article on Simple-Talk that explains a bunch of them and shows the performance differences between them. Here...

You're also looking at table scans from this bit of code:

            
March 22, 2017 at 5:49 am
            
#1934275

            
            
                                                
        

Your only hope is that the application name is supplied with the connection. It's optional, so you may not see it. I'd strongly suggest, as has everyone else, cleaning up...

I'd suggest setting up trace events to capture remote procedure calls and sql batch. You should be able to identify the login from there. The application name might not be...

You mean the extended events for Idera are missing? Not sure. You might want to see if Idera has a forum or some means of asking this question.

Since you can't go to 2016 and take advantage of Query Store, upgrading to 2014, you need to, as much as possible, simulate having Query Store available. With that in...

Excellent article.

I would suggest a single caveat that having a secondary copy of the data somewhere other than Azure is a good practice if this is business critical...

It's not always about free consulting. When we had people get past our tier 1 phone interview, assuming they did well during our second discussion (white board & all that...

Interesting. I've never seen that approach before. I'm not sure I'm crazy about it at all. In general I wouldn't really bother counting, or worrying about the count, of CXPACKETS....

I'm not sure where you got the concept of balancing CXPACKET and SOS_SCHEDULER_YIELD. That's not a valid approach that I've ever heard of.

CXPACKETS just indicates that parallelism is...

However, I'm a little hung on why this is so vitally important? What's the end goal? What are we trying to do? Is it just about learning index internals?