It's filtering 14 days here:

WHERE exec_date >= dateadd( day, -14, getdate() )

So, change the -14 to -1 and you're pretty much finished if everything else is working for you.

I don't...

But you can create a nonclustered columnstore with a row-based clustered index.

I'd still run a consistency check.

Rebuilding an index updates the statistics at the same time. So the order in which you run these things does matter. Understand that if an index is not rebuilt, it...

After reading through all this excellent advice, the only thing I'd add is a question. Do you really need DISTINCT? That's adding quite a lot of overhead on top of...

I like what Jason did. As soon as you start seeing correlated sub-queries like that, usually the best approach is either a JOIN or an APPLY.

Sean Lange (8/19/2015)

---

I am working on putting together my first SSIS package. I haven't done this type of thing since it was still DTS packages. This stuff is a LOT...

If you hit problems with the book, you can always post questions here.

Here's the documentation on how to use ATTACH.

BL0B_EATER (8/14/2015)

---

Grant Fritchey (8/13/2015)

---

Eric M Russell (8/13/2015)

---

SQL Azure should be technically up to the task. However, my understanding is that you get billed based on network transfer fees. How many...

The fact that it ran differently on your machine could very easily be because it compiled a new, and better, execution plan on your machine. Also, what about blocking and...

They should be. They don't have to be. I'd sure put them up there. You'll pay for those VMs, but those costs should, by and large, be less than moving...

Eric M Russell (8/13/2015)

---

SQL Azure should be technically up to the task. However, my understanding is that you get billed based on network transfer fees. How many GB total of...

That makes absolute sense.

Interesting. Then it prefers to use an average of the statistics and a generic plan, not a specific plan based on a value. Sounds like it might be bad parameter...