-Is there a way to keep all intermediate files there after the MapReduce Job run?
In order to know more about the files creation and size when the job is
running, I want to keep all the intermediate files there (job.xml,
spillN.out, file.out, file.index, map.out-N, etc).
My question is :
1. Is there any configurations that can make this happen? Or could I modify
some Hadoop MapReduce code for this ?
2. Since each job, each task, and each attempt of the task using different
directories to store all the intermediate files, keeping the files there
without deleting will not hurt the whole MapReduce cluster except taking up
some storage. Am I write?
Michael Segel 2013-03-01, 13:23
Jean-Marc Spaggiari 2013-03-01, 13:49