AnsweredAssumed Answered

Setting mapred.max.split.size value

Question asked by sambaiahk on Dec 3, 2013
Latest reply on Dec 5, 2013 by sambaiahk
Hi,
  I am running examples as hadoop jar hadoop-0.20.2-dev-examples.jar  wordcount -D mapred.job.queue.name=myqueue -D mapred.max.split.size=10000000000 input output

I have only one file under input directory of size 6268149090 so I expect to run single map task. Instead mapr is starting 24 map tasks that means max.split.size is 256MB. I wanted to run single map task instead of 24 map tasks. What is that I am missing?

Outcomes