Shawn Melton (11/13/2014)
I am not sure what the actual question is here but if you are referring to David Klee's presentation you can re-watch it on PassTV here
The main point I think he is trying to make is to monitor your metrics of that virtual machine and host machine to make sure you sized your VM resources properly for a SQL Server instance. There is no one size fits all scenario to go by as is more common to do.
Try to run this query from MS-SQL:
It shows you how many nodes you have and how many cores per node.
What I am trying to establish is the performance correlation, same SQL instance, between nodes and core. And let me give you an example:
-6 nodes with 1 core per node
-3 nodes with 2 cores per node
-2 nodes with 3 cores per node
... all that on same exact VM/SQL server.
I want to hear from people, real experience, if playing around on VMware this way, they saw CPU performance improvements. When doing the math, they all have same number of cores, six in total, but VM configuration is different.
Yes, I understand and I know that the "one size fits all" mentality is wrong :-) ... I am no questioning that.