I have experience with hadoop 0.20.204 on 3 machines cluster as pilot,
now im trying to setup real cluster on 32 linux machines.
I have some question:
1. is hadoop 1.0 stable?? in hadoop site this version is indicated as
2. as you know installing and setting up hadoop in all 32 machines
separately in not good idea, so what can i do?
1. using hadoop on demand (HOD)?
2. or using OS image replicate tools same as clozeZilla? i think
this method is better because in addition to hadoop I can clone
same other settings such as SSH or Samba in all machines.
Let me know your idea,