Jeff Moden (2/9/2013)

---

Freakin' awesome link, Paul. I was able to modify Barry's code to also correctly populate the MX column for the Logoffs so that I could graph the...

As far as optimizing the original query is concerned, I would be interested to see the execution plans and performance information for:

CREATE TABLE #Temp

(

TranID integer...

MyDoggieJessie (2/7/2013)

---

When I go to actually use the data from the CTEs and issue an UPDATE using data from it, it takes nearly 3 minutes to run and returns millions...

drew.allen (2/8/2013)

---

You're not using the Txn CTE anywhere in your UPDATE statement, so you can drop it without changing the results of your query. I believe that is the...

It's possible that using CRYPT_GEN_RANDOM in a function was (incorrectly) possible...

Kwisatz78 (2/8/2013)

---

I have a table which contains login and logout times for a large set of users, and we are wanting to know how to code it to pull...

Lynn Pettis (2/8/2013)

---

At this point, I will bow out and defer to Paul. He has spent much more time than I digging into these things than I and has...

Roger Sabin (2/8/2013)

---

I am having a hard time understanding why the first index would be used when it requires tremendously more IO and time than the second index to return...

+1 to what Tom said.

L' Eomot Inversé (12/31/2012)

---

For some reason the link doesn't work. But looking in the obvious place indicated by your link (http://research.microsoft.com/en-us/um/people/jrzhou/publications.htm) finds it and the link from there, although...

Dwain,

I see rather different performance using your test rig on SQL Server 2012 (build 3321) and SQL Server 2008 R2 (build 4266):

+--------------------------------------------------+

¦ SQL Server ¦ PatternSplitCM ¦ SQL#.RegEx_Matches ¦

¦------------+----------------+--------------------¦

¦ 2012...

L' Eomot Inversé (12/30/2012)

---

Why can't the optimiser look and see if spooling would enhance performance (only in the cases where it make no semantic difference, of course) and use it...

Another good one. 19% wrong right now :blink:

Excellent one-point question. 21% wrong answers = no such thing as a too-easy QotD.

L' Eomot Inversé (12/17/2012)

---

Hugo Kornelis (12/17/2012)

---

(That being said - there is indeed no guarantee. SQL Server could come up with an execution plan that projects the SELECT list first, then...