How do you control nodes in a server farm?

Question

I've been reading about hadoop and multi-node setups, and it says in the documentation that you must have a JVM and hadoop software already running on those nodes.

My question is, do people install this software on each of these computers individually? or is there a software solution that automates this process?

I've also read about KVM switches, but I'm not sure whether this is what people usually use in these situations.