Method to schedule the number of Mappers and Reducers in a Hadoop MapReduce Tsk

Question

Am trying to Schedule a MapReduce job where in which I had programmed mapper tasks to a limited number of 20 and on the other hand I had Programmed the Reducer Tasks to 0 but, Still, I ended up at getting a value other than zero. The system is unc=successful in displaying the time consumed to perform the task and am unable to understand where am I going wrong. Can anyone help me?

Answer 1

The number of Mapper tasks for a particular job is always equal to the number of input splits which means for each and every input split a Mapper task gets deposited. so the bottom line is that the entire cycle of MapReduce job might have a certain number of input splits to which they might have got themselves the equal number of Mapper tasks.

In your case, You might have input splits exceeding 20 to which you might have got the same number of Mapper Tasks 

You can get more detailed information on this link.