1.TestDFSIO
Enter the directory where the Jar package is located:
# cd /usr/local/src/hadoop/hadoop-2.7.7/share/hadoop/mapreduce/
# ls
(1) Write
Write 5 128MB files to HDFS
# hadoop jar hadoop-mapreduce-client-jobclient-2.7.7.jar TestDFSIO -write -nrFiles 5 -fileSize 128MB
(2) Read
Read 5 128MB files from HDFS
# hadoop jar hadoop-mapreduce-client-jobclient-2.7.7.jar TestDFSIO -read -nrFiles 5 -fileSize 128MB
(3) View
# cat TestDFSIO_results.log
(4) Clear
# hadoop jar hadoop-mapreduce-client-jobclient-2.7.7.jar TestDFSIO -clean
2.mrbench
Use 3 mappers and 3 reducers to run a small job 5 times, generate 5 input rows, sort in descending order
# hadoop jar hadoop-mapreduce-client-jobclient-2.7.7-tests.jar mrbench -numRuns 5 -maps 3 -reduces 3 -inputLines 5 -inputType descending
[Job 1]
[Job 2]
[Job 3]
[Job 4]
[Job 5]
[Operation results]
3.nnbench
Use 3 mappers and 3 reducers to create 5 files
# hadoop jar hadoop-mapreduce-client-jobclient-2.7.7-tests.jar nnbench -operation create_write -maps 3 -reduces 3 -numberOfFiles 5 -replicationFactorPerFile 3 -readFileAfterOpen true
