Given the phrasing of the question, I guess you're...

SQL 2014, I assume, since you mention clustered columnstore.

The clustered columnstore is a clustered index, meaning it is the primary storage of the table, hence add columns to the table...

The memory-optimised tables are memory-resident, they cannot not be in memory. What memory is left over from that is the normal SQL buffer pool and managed as normal for the...

That can happen if there are inserts into the same table on both nodes at the same time. There's a latency on the replication.

This is why the design guidance for...

It might be in the default trace. If not, then unlikely unless you had some custom monitoring in place.

Estimated plan can be perfectly adequate. Only time you need the actual plan is if you're looking at the row estimates and whether they're wrong. For just finding table/index scans...

And if db_datareader didn't 'work', then manually adding exactly the...

That query generates the statement, copy-paste once you've checked it, then run it.

I'll echo, use roles. Directly assigning users to objects causes major admin overhead and debugging pain. db_datareader gives...

Yup, bad design will almost certainly result in sub-standard performance

Your comments...

If you want accurate results, you probably don't want to use a script written for SQL 2000 on SQL 2014. Find a correct one for recent versions of SQL (ie...

andymay-496937 (6/19/2016)

---

as the clustered index key is always added to any non-clustered indexes meaning that 2 look-ups are always required to find rows, the first to get the collection of...

Depending how the 'table size' is being calculated, it may not take the LOB columns into account.

That's probably why. Varchar(max) columns are stored out of row, so depending how that 'table size' is being calculated, they...

Don't know what happened to the other thread...

Was that file in a secondary filegroup, or was it part of Primary?

You...