利用hadoop自带程序运行wordcount

1.启动hadoop守护进程

bin/start-all.sh

2.在hadoop的bin目录下建立一个input文件夹

JIAS-MacBook-Pro:hadoop-0.20. jia$ mkdir input

3.进入input目录之后，在input目录下新建两个文本文件，并想其写入内容

JIAS-MacBook-Pro:hadoop-0.20. jia$ cd input
JIAS-MacBook-Pro:input jia$ echo "hello excuse me fine thank you">text1.txt
JIAS-MacBook-Pro:input jia$ echo "hello how do you do thank you">text2.txt

4.进入hadoop的bin目录，输入jps命令，确认hadoop已经跑起来了

JIAS-MacBook-Pro:hadoop-0.20. jia$ cd bin
JIAS-MacBook-Pro:bin jia$ jps
 SecondaryNameNode
 NameNode
 JobTracker
 TaskTracker
 DataNode
 Jps

5.把input文件上传到hdfs上

JIAS-MacBook-Pro:hadoop-0.20.2 jia$ bin/hadoop dfs -put input in

6.查看hdfs上的项目

JIAS-MacBook-Pro:hadoop-0.20. jia$ bin/hadoop dfs -ls ./in/*
-rw-r--r--   1 jia supergroup         31 2014-07-17 20:39 /user/jia/in/text1.txt
-rw-r--r--   1 jia supergroup         30 2014-07-17 20:39 /user/jia/in/text2.txt

7.利用自带的wordcount执行，并把结果放在output文件夹上

JIAS-MacBook-Pro:hadoop-0.20. jia$ bin/hadoop jar hadoop-0.20.-examples.jar wordcount in output
// :: INFO input.FileInputFormat: Total input paths to process :
// :: INFO mapred.JobClient: Running job: job_201407172036_0001
// :: INFO mapred.JobClient:  map % reduce %
// :: INFO mapred.JobClient:  map % reduce %
// :: INFO mapred.JobClient:  map % reduce %
// :: INFO mapred.JobClient: Job complete: job_201407172036_0001
// :: INFO mapred.JobClient: Counters:
// :: INFO mapred.JobClient:   Map-Reduce Framework
// :: INFO mapred.JobClient:     Combine output records=
// :: INFO mapred.JobClient:     Spilled Records=
// :: INFO mapred.JobClient:     Reduce input records=
// :: INFO mapred.JobClient:     Reduce output records=
// :: INFO mapred.JobClient:     Map input records=
// :: INFO mapred.JobClient:     Map output records=
// :: INFO mapred.JobClient:     Map output bytes=
// :: INFO mapred.JobClient:     Reduce shuffle bytes=
// :: INFO mapred.JobClient:     Combine input records=
// :: INFO mapred.JobClient:     Reduce input groups=
// :: INFO mapred.JobClient:   FileSystemCounters
// :: INFO mapred.JobClient:     HDFS_BYTES_READ=
// :: INFO mapred.JobClient:     FILE_BYTES_WRITTEN=
// :: INFO mapred.JobClient:     FILE_BYTES_READ=
// :: INFO mapred.JobClient:     HDFS_BYTES_WRITTEN=
// :: INFO mapred.JobClient:   Job Counters
// :: INFO mapred.JobClient:     Launched map tasks=
// :: INFO mapred.JobClient:     Launched reduce tasks=
// :: INFO mapred.JobClient:     Data-local map tasks=
JIAS-MacBook-Pro:hadoop-0.20. jia$ 

8.查看结果

JIAS-MacBook-Pro:hadoop-0.20. jia$ bin/hadoop dfs -ls
Found  items
drwxr-xr-x   - jia supergroup           -- : /user/jia/in
drwxr-xr-x   - jia supergroup           -- : /user/jia/output
JIAS-MacBook-Pro:hadoop-0.20. jia$ bin/hadoop dfs -ls ./output
Found  items
drwxr-xr-x   - jia supergroup           -- : /user/jia/output/_logs
-rw-r--r--    jia supergroup          -- : /user/jia/output/part-r-
JIAS-MacBook-Pro:hadoop-0.20. jia$ bin/hadoop dfs -cat ./output/*
do    2
excuse    1
fine    1
hello    2
how    1
me    1
thank    2
you    3
cat: Source must be a file.