SQL Azure is the latest major investment in database scaling technology. There is also upgraded peer replication that adds geo-distributed synchronized sql database capability as well.

I believe the implication is that Hibernate, which is a RedGate product (and they are known for their performance oriented applications), is more efficient with its memory management than SSIS,...

It just occurred to me that you should use a CTE...select the rows from tha small tables. This might keep everything in memory..not sure.

Then, with multiple connection/update interfaces, you are stuck with triggers which evaluate every update to determine whether it should be allowed or not.

Hmm...maybe with a snapshot you could let...

The duration of the 'outage' on any Windows cluster is the sum of the time it takes to detect a failover is required plus the time it takes to unmount...

Adding a second instance with separate resources to a cluster is a standard operation so...to the last post...nothing unusual or difficult.

To the original post...your end configuration would be exactly the...

Stopping the cluster service on the active node is probably as gentle a method as you will find. You can do that via the cluster node /stop command or a...

To second GilaMonster's comment...we cannot exempt ourselves as DBAs from the rules we apply to Developers...ALWAYS have a rollbadk plan...NEVER make changes without protection.

You can also use the CAST statement since you wanted a DATE data type...not VARCHAR with an implicit conversion

cast(<datetime column> as date) as DateOnly

It looks like it has to be deployed on a SQL 2008 instance. The collection sets include scheduled SSIS packages which access objects like dm views so there is a...

It is very possible he is right. Set up representative tests and prove it one way or the other. Be sure to include development effort and package creation in the...

Can you split your data into two tables with a UNION view over both? One table, for 'open' items can be updated, the other for 'closed' items has an INSTEAD...

This could be the result of a very light workload and very small number of queries executing on the test server...not enough memory pressure to force the query plan and...

I have no experience with this but here are my thoughts..not in any particular order...

First...network bandwidth. You have not stated whether the servers are in the same physical location or...

You can make something of an educated guess if, over time, you establish a relationship between 'query cost' as reported in the execution plan, and the actual time it took...