I want to test version hadoop-0.22.
But when configurate the fairescheduler, I have some troublesome. The
fairscheduler is not efficient.
And I have configured this items in the mapred-site.xml, and also I hava
copy the fairscheduler jar file to the $HADOOP_HOME/lib:
mapreduce.jobtracker.ta
It works now if the number of reducers is limited (4 in my case).
However I am not sure why sometimes it doesn't work if the number of
reducers is increased? I tried to increase the number of open files
for the user as suggested by some blog, but still for large number of
reducers it doesn't work.
