To me, splitting it is the trivial part. Analyzing all the possible variations will be vastly more complex.

If possible, I'd just add code to that job to write its status to a table in the other db. That way you could avoid referencing msdb at all...

I doubt you'll be able to recover the full number. Floats are imprecise beyond their guaranteed number of digits. You'll likely have to re-export the data.

That said, you...

andrew gothard (4/2/2014)

---

ScottPletcher (3/30/2014)

---

andrew gothard (3/29/2014)

---

Simple arithmetic, if you have 8k pages, one row per page (and we know that it's going to be bigger than that), 12500 rows. ...

AZ Pete (4/2/2014)

---

Thanks folks for the replies. However, my question is *why* did SQL Server elect to use the unique index defined on the TransmissionId column versus using the Primary...

To allow the best possible use of any existing indexes, approach 1 is vastly superior to 2. Neither is best, however. Instead, assuming "endDate" is a date/datetime/etc., use...

I think this will depend most on your I/O subsystem's speed.

But if there's only 30G total worth of data, I think you'd best off just doing a full shrink:

DBCC SHRINKFILE...

For DBA, be aware that having no experience will make this a very difficult transition. Most employers don't hire completely inexperienced DBAs.

For DBA, download SQL Express, 2008 or 2012.

If...

Ed Wagner (3/31/2014)

---

I've found the creation of a script and then restoring from a production backup to be very reliable.

ScottPletcher (3/31/2014)

---

Of course, if it's possible, it's much easier longer-term to...

New Born DBA (3/31/2014)

---

ScottPletcher (3/31/2014)

---

Yes, restore the qa db from a production backup. Create a script that does that, including all the appropriate "WITH MOVE ... TO ..., MOVE...

Yes, restore the qa db from a production backup. Create a script that does that, including all the appropriate "WITH MOVE ... TO ..., MOVE ... TO ..." clauses.

But...

andrew gothard (3/29/2014)

---

Simple arithmetic, if you have 8k pages, one row per page (and we know that it's going to be bigger than that), 12500 rows. Table Lock (if...

Ignoring any potential performance issues, a CROSS APPLY can do what you want fairly easily. Simply indexing the table properly can often deal with that. Or you can...

SELECT

COALESCE(a.ID, b.ID) AS ID,

COALESCE(a.NAME, b.NAME) AS NAME,

CASE WHEN a.UpdateDate > b.UpdateDate THEN a.UpdateDate ELSE b.UpdateDate END...

andrew gothard (3/27/2014)

---

ScottPletcher (3/27/2014)

---

andrew gothard (3/27/2014)

---

ScottPletcher (3/27/2014)

---

yuvipoy (3/27/2014)

---

Sean Lange (3/27/2014)

---

True. I have worked with this OP on several threads and the problem is performance because they have several related tables...