Default Mapper-Reducer class

Question

Suppose I have two dataset :

hello world
bye world

and

hello earth
new earth

and I want to run a map-reduce task which does not specify mapper class or reducer class, So the default mapper and reducer will be called - which both are identity function. When I run the job the output is ::

0       hello world
0       hello earth
12      new earth
12      bye world

I am confused why the key is like 0 and 12 ?! I just used default mapper and reducer as I commented out these lines in the main() ::

//    job.setMapperClass(Map.class);
//    job.setCombinerClass(Reduce.class);
//    job.setReducerClass(Reduce.class);

So, my question is what is the output key is here ? why it looks like 0, 0 , 12, 12 ?

Answer 1

0,0,12 and 12 are the file offsets in the input data. In the case of text inputs the K to the mapper is the file offset and the value is the input line. Check this for more information.