Hello guys, I hope my question does not sound weird, but could a Spark execution on Hadoop cluster give different output than the program actually does ? I mean by that, the execution order is messed by hadoop, or an instruction executed twice..; ?
Thanks for your enlightenment
