Cluster shared volumes

The most common question on cluster shared volumes is how many CSVs you need and how huge they may get when filled with data. As mentioned before, it's a good rule of thumb to create one CSV per cluster node; in larger environments with more than eight cluster nodes, a CSV per two to four nodes. The number of VMs per-CSV is not limited. Commonly, I do not see more than 50 VMs on a CSV for server VMs and 100 VMs for client VMs in the VDI environment. However, don't think in units here, plan in IOPS. Spread the IOPS evenly between your CSVs. To utilize the redundant storage hardware I wrote about earlier, never use a single CSV. Start with at least two CSVs to spread the load over your two storage controllers. This isn't necessarily a design of CSVs, rather a behavior of the SAN and how it manages its disks. If you use one CSV, it's possible that the SAN allocates ownership of that LUN to a single controller and could introduce a bottleneck in performance. Dividing the...