I run MapReduce as hdfs user and it is executed successfully only thing is that I got generated 15
part-r-000** files from which only 2 contain some results (those 2 files contain all the result which is needed but 13 files are with 0 bytes size). So why these files are generated? And also since two files
part-r-00012 are quite small we would like to have all the results in one files.
In Cloudera when we were running the same thing, we got only one file with same results.
So how we can achieve this?
I am familiar with getMerge command - but this should not be the resolution of this case, because multiple part-***** files with size 0 bytes should not be created at all.