My experience broadly echoes Glenn's recommendation; off for OLAP, on for OLTP. But, as you mention in the article no two workloads are ever the same.

These sorts of scenarios...

I've done both in the past.

In some organisations we've added a 'SystemDB' which is used for general purpose logging, supplementary backup info, capturing occasional profiler traces etc. For views...

Table variables are not inherently slower than temp tables. The issue lies in the optimiser evaluating them as always having a small number of rows, IIRC < 10. That aside,...

You're getting various inappropriate and/or plain inaccurate suggestions thrown in here.

No. There can be circumstances where you...

john 64806 (10/17/2010)

---

Pages / Sec refers to how much memory is addressed, it's like saying "you're doing so and so many reads per second to disk" but about memory instead,...

The common table expressions used are an alternative to using temporary or derived tables, hopefully making the query easier to understand by breaking it down into smaller parts. Hastily put...

This tweaked my interest on the train journey home this evening. Extension/merge of the two previous suggestions, using the script to generate data that ColdCoffee kindly added but using perm...

Exactly the same issue (and bizarrely almost identical counts from Profiler and QA) cropped up here recently. I've been scratching my head and getting no where so far.

Apparent blocking problems post SP4 are bound to be http://support.microsoft.com/default.aspx?scid=KB;EN-US;906344

While you should generally run the tuning wizard as a matter of course as the first step with performance problems, you have to remember that it can only tell you...

If there isn't a policy to stop 'really bad' articles being published on SQLServerCentral, there should at least be one that stops said articles appearing in the "One year ago..."...

Have no issue with the advertisements myself, you're perfectly entitled to do whatever is necessary to earn a crust! Would however agree that they may work better on the opposite...

Tom mentioned it but I didn't see a response, have you checked the execution plans on both servers to see if there are any differences between the servers?

I'm starting to wonder if some of the people in the "DRI is stupid" camp are mis-understanding what is meant by referential integrity. Some posters in the original thread seemed...

Litespeed and Sonasafe are both worth investigating.