Hi there!
I've been managing a few ESXi hosts across multiple data centers and we're starting to look into options for consolidating the VMs onto shared storage to benefit from vMotion and HA.
From reading the communities and various sources around the 'net I have learned that a good indication of IOPS done by the ESXi can be calculated from the Disk performance graph using the Read and Write requests, then dividing the values by the graph periods (20s in my case).
Now comes the question: should I sum all the IOPS values of each ESXi host that is going to be connected (and using) the SAN or is there a different way of calculating that?
I'm calculating IOPS always using the MAX values of each graph.
