java - Profiling hadoop job with yourkit -
i following tutorial http://www.michael-noll.com/tutorials/running-hadoop-on-ubuntu-linux-single-node-cluster
i want see method or java process use cpu wordcount mapper going through corpus using yourkit. understand hadoop type "text" in utf-8 , java string in utf-16. trying see wordcount mapper work in converting utf-8 utf-16.
however, yourkit doesn't show details process. shows cpu time follow: org.apache.hadoop.mapred.child.main(string[]) ~ 96% , org.apache.hadoop.hdfs.dfsclient$dfsoutputstream$responseprocessor.run() ~ 4%
i not familiar yourkit. please point out how should approach this?
Comments
Post a Comment