Hadoop 实战之Streaming（一）

环境：Vmware 8.0 和ubuntu11.04

Hadoop 实战之Streaming（一）---通过Unix命令使用Streaming

第一步: 首先在/home/tanglg1987目录下新建一个start.sh脚本文件，每次启动虚拟机都要删除/tmp目录下的全部文件，重新格式化namenode，代码如下：

sudo rm -rf /tmp/*
rm -rf /home/tanglg1987/hadoop-0.20.2/logs
hadoop namenode -format
hadoop datanode -format
start-all.sh
hadoop fs -mkdir input 
hadoop dfsadmin -safemode leave

第二步：给start.sh增加执行权限并启动hadoop伪分布式集群，代码如下:

chmod 777 /home/tanglg1987/start.sh
./start.sh 

执行过程如下：

第三步：编写一个名为：list-4-1.sh的shell脚本

$HADOOP_HOME/bin/hadoop  jar $HADOOP_HOME/hadoop-0.20.2-streaming.jar \
    -input input/cite75_99.txt \
    -output output \
    -mapper 'cut -f 2 -d ,' \
   -reducer  'uniq'

第四步：给list-4-1.sh增加执行权限并启动脚本，代码如下：

chmod 777 /home/tanglg1987/list-4-1.sh
./list-4-1.sh

第五步：Run On Hadoop，运行过程如下：

tanglg1987@tanglg1987:~/test/streaming$ ./list-4-1.sh
packageJobJar: [/tmp/hadoop-tanglg1987/hadoop-unjar6010847675622454264/] [] /tmp/streamjob1903205486478029700.jar tmpDir=null
12/10/19 21:07:00 INFO mapred.FileInputFormat: Total input paths to process : 1
12/10/19 21:07:00 INFO streaming.StreamJob: getLocalDirs(): [/tmp/hadoop-tanglg1987/mapred/local]
12/10/19 21:07:00 INFO streaming.StreamJob: Running job: job_201210192029_0004
12/10/19 21:07:00 INFO streaming.StreamJob: To kill this job, run:
12/10/19 21:07:00 INFO streaming.StreamJob: /home/tanglg1987/hadoop-0.20.2/bin/../bin/hadoop job -Dmapred.job.tracker=localhost:9101 -kill job_201210192029_0004
12/10/19 21:07:00 INFO streaming.StreamJob: Tracking URL: http://localhost:50030/jobdetails.jsp?jobid=job_201210192029_0004
12/10/19 21:07:01 INFO streaming.StreamJob: map 0% reduce 0%
12/10/19 21:07:16 INFO streaming.StreamJob: map 7% reduce 0%
12/10/19 21:07:19 INFO streaming.StreamJob: map 10% reduce 0%
12/10/19 21:07:22 INFO streaming.StreamJob: map 15% reduce 0%
12/10/19 21:07:25 INFO streaming.StreamJob: map 19% reduce 0%
12/10/19 21:07:28 INFO streaming.StreamJob: map 23% reduce 0%
12/10/19 21:07:31 INFO streaming.StreamJob: map 26% reduce 0%
12/10/19 21:07:34 INFO streaming.StreamJob: map 29% reduce 0%
12/10/19 21:07:37 INFO streaming.StreamJob: map 33% reduce 0%
12/10/19 21:07:40 INFO streaming.StreamJob: map 36% reduce 0%
12/10/19 21:07:43 INFO streaming.StreamJob: map 40% reduce 0%
12/10/19 21:07:46 INFO streaming.StreamJob: map 45% reduce 0%
12/10/19 21:07:49 INFO streaming.StreamJob: map 48% reduce 0%
12/10/19 21:07:52 INFO streaming.StreamJob: map 50% reduce 0%
12/10/19 21:08:19 INFO streaming.StreamJob: map 58% reduce 0%
12/10/19 21:08:22 INFO streaming.StreamJob: map 58% reduce 17%
12/10/19 21:08:25 INFO streaming.StreamJob: map 63% reduce 17%
12/10/19 21:08:28 INFO streaming.StreamJob: map 67% reduce 17%
12/10/19 21:08:31 INFO streaming.StreamJob: map 71% reduce 17%
12/10/19 21:08:34 INFO streaming.StreamJob: map 74% reduce 17%
12/10/19 21:08:37 INFO streaming.StreamJob: map 77% reduce 17%
12/10/19 21:08:40 INFO streaming.StreamJob: map 81% reduce 17%
12/10/19 21:08:43 INFO streaming.StreamJob: map 84% reduce 17%
12/10/19 21:08:46 INFO streaming.StreamJob: map 88% reduce 17%
12/10/19 21:08:49 INFO streaming.StreamJob: map 91% reduce 17%
12/10/19 21:08:52 INFO streaming.StreamJob: map 95% reduce 17%
12/10/19 21:08:55 INFO streaming.StreamJob: map 97% reduce 17%
12/10/19 21:08:58 INFO streaming.StreamJob: map 99% reduce 17%
12/10/19 21:09:02 INFO streaming.StreamJob: map 100% reduce 17%
12/10/19 21:09:25 INFO streaming.StreamJob: map 100% reduce 33%
12/10/19 21:09:28 INFO streaming.StreamJob: map 100% reduce 70%
12/10/19 21:09:31 INFO streaming.StreamJob: map 100% reduce 73%
12/10/19 21:09:34 INFO streaming.StreamJob: map 100% reduce 79%
12/10/19 21:09:37 INFO streaming.StreamJob: map 100% reduce 84%
12/10/19 21:09:40 INFO streaming.StreamJob: map 100% reduce 90%
12/10/19 21:09:43 INFO streaming.StreamJob: map 100% reduce 95%
12/10/19 21:09:49 INFO streaming.StreamJob: map 100% reduce 100%
12/10/19 21:09:55 INFO streaming.StreamJob: Job complete: job_201210192029_0004
12/10/19 21:09:55 INFO streaming.StreamJob: Output: output

第六步：查看结果集，运行结果如下：