Nonclustered indexes are logically sorted by the index key. What are you trying to do?

In general, the wait statistics before and after the slow time to see what's causing things to run slow during the slow time. If you know the specific queries that...

Since you're updating the table, and the table is the clustered index, no matter what, the cluster is going to be accessed. However, look at the execution plan. If it's...

I suspect you have something else going on there. Those statements are SELECT statements. They're not going to be blocking in this fashion. Instead, there must be something taking out...

Normally to set up a reporting system you'd look to something like Availability Groups (where you can set up a read_only secondary), log shipping, mirroring, or even replication. These are...

helper_10 (3/19/2016)

---

There are couple of ways to do that; First you can download AdventureWorks Database file with .Bak extension and use restore method to restore to your Management Studio. Secondly,...

I've never heard of such a thing, no. Unless you've found some kind of very obscure bug, I can almost 100% guarantee that something is being done on that system....

I'm getting here late, but I'm with Gail and Jeff. It sure sounds like the problem is primarily one of query tuning and possibly the cost threshold for parallelism. Change...

joshdbguy (3/15/2016)

---

Grant Fritchey (3/15/2016)

---

Not enough info to go on. Trace, or extended events, is the way to go to capture the queries that are being called. The real question is,...

If you're designing the database, focus first and foremost on the clustered index. Make sure that it is the right choice (and big hint, the primary key is not always...

Not enough info to go on. Trace, or extended events, is the way to go to capture the queries that are being called. The real question is, is the server...

What the other people have said...

CXPacket is only a concern if you also see other CPU waits. The one thing I will say, sight unseen, if your Cost Threshold for...

I doubt there's a document to support that. It's a bad estimate. I've successfully run 200gb databases on servers with 16gb of memory. It really depends on the load on...

Andy Warren (3/14/2016)

---

Grant, in my view what has slowed adoption is how long its taken to get to a "real" version of SQL in the cloud (arguably v12). Before that...

Gary Varga (3/14/2016)

---

Grant Fritchey (3/14/2016)

---

I've seen...