黑猴子的家：Hadoop 本地模式运行案例

1、官方grep案例

1）在hadoop-2.8.2文件下面创建一个input文件夹

[victor@node1 hadoop-2.8.2]$ pwd
/opt/module/hadoop-2.8.2
[victor@node1 hadoop-2.8.2]$ mkdir input

2）将hadoop的xml配置文件复制到input

[victor@node1 hadoop-2.8.2]$ pwd
/opt/module/hadoop-2.8.2
[victor@node1 hadoop-2.8.2]$ cp -r etc/hadoop/*.xml input/

3）执行share目录下的mapreduce程序

[victor@node1 hadoop-2.8.2]$ bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.8.2.jar grep input output 'dfs[a-z.]+'
[victor@node1 hadoop-2.8.2]$ cat output/* 
1       dfsadmin
注意：output 不存在，不用提前创建

2、官方wordcount案例

1）在hadoop-2.8.2文件下面创建一个wcinput文件夹

[victor@node1 hadoop-2.8.2]$ pwd
/opt/module/hadoop-2.8.2
[victor@node1 hadoop-2.8.2]$ mkdir wcinput

2）在wcinput文件下创建一个wc.input文件

[victor@node1 hadoop-2.8.2]$ cd wcinput/
[victor@node1 wcinput]$ > wc.input
[victor@node1 wcinput]$ ls
wc.input

3）编辑wc.input文件

[victor@node1 wcinput]$ vi wc.input 
hadoop hdfs yarn mapreduce
dfs hdfs hdfs yarn English victor
spark haha
ip id id nat ict mat cctv cctv
保存退出  :wq

4）执行案列

[victor@node1 wcinput]$ cd ../
[victor@node1 hadoop-2.8.2]$ pwd
/opt/module/hadoop-2.8.2
#[victor@node1 hadoop-2.8.2]$ bin/hadoop fs -put wcinput/ /user/victor/wcinput
[victor@node1 hadoop-2.8.2]$ bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.8.2.jar wordcount wcinput wcoutput
注意：wcoutput 不存在，不用提前创建，本地模式不用上传到hdfs上

5）查看结果

[victor@node1 hadoop-2.8.2]$ cat wcoutput/*
English 1
cctv    2
dfs     1
hadoop  1
haha    1
hdfs    3
ict     1
id      2
ip      1
mapreduce   1
mat     1
nat     1
victor  1
yarn    2
spark  1