The default behavior is that the input splits are where the data is stored.
On 4/4/12 5:24 PM, "sdnetwork" <sdnetw...@gmail.com> wrote: >ok thanks, > >but i don't find the information that tell me how the result of the split >is >distrubuted across the different node of the cluster ? > >1) randomely ? >2) where the data is stored ? > > > > >