Recommandations for fast OS cloning on cluster

[John Lilley]

Are there standard approaches for setting up a Hadoop cluster "quickly" by 
cloning most of the data nodes?