Cluster Shared Volumes (CSV) is a new feature introduced in Windows Server 2008 R2. After you’ve setup your cluster, enabled CSV, and added clustered disks to CSV, its now time to copy your VHD files to those CSV disk. The big question is, which cluster node should you perform this operation on? Beauty of CSV is that because all nodes can access CSV volume, you can open Windows Explorer and browse to C:\ClusterStorage from any node to see VHD file. So now you’re ready to copy multiple VHDs to CSV volumes in your cluster. If you’re deploying many cloned VHDs, you may be doing this operation many times and copying hundreds or thousands of GB of data. But, should you do that from any node? Did you consider that choice might make a difference in how long operation can take?
Full Article: MSDN blog