For the virtualized workloads, the data is already there and can be used for the design (we will discuss later how to get and use it); maybe the greater complexity is to be found with other hypervisors, where host-related metrics could not be easily comparable with a vSphere environment. However, data collected at VM level could be good enough and quite useful.
Monitoring could be handled in the same way as with the physical environment, with the same tools and the same metrics, but it's easy to gain more information directly from your vCenter Server without having to deal with every single guest operating system. Monitoring and collecting (performance) data from a virtual environment will be discussed in depth in Chapter 9, Monitoring, Optimizing, and Troubleshooting.
But it's not all about metrics; you may also need to...