運行Hadoop自帶的wordcount單詞統(tǒng)計程序

0.前言

    前面一篇《Hadoop初體驗：快速搭建Hadoop偽分布式環(huán)境》搭建了一個Hadoop的環(huán)境，現(xiàn)在就使用Hadoop自帶的wordcount程序來做單詞統(tǒng)計的案例。

1.使用示例程序?qū)崿F(xiàn)單詞統(tǒng)計

（1）wordcount程序

    wordcount程序在hadoop的share目錄下，如下：

[root@leaf mapreduce]# pwd
/usr/local/hadoop/share/hadoop/mapreduce
[root@leaf mapreduce]# ls
hadoop-mapreduce-client-app-2.6.5.jar         hadoop-mapreduce-client-jobclient-2.6.5-tests.jar
hadoop-mapreduce-client-common-2.6.5.jar      hadoop-mapreduce-client-shuffle-2.6.5.jar
hadoop-mapreduce-client-core-2.6.5.jar        hadoop-mapreduce-examples-2.6.5.jar
hadoop-mapreduce-client-hs-2.6.5.jar          lib
hadoop-mapreduce-client-hs-plugins-2.6.5.jar  lib-examples
hadoop-mapreduce-client-jobclient-2.6.5.jar   sources

    就是這個hadoop-mapreduce-examples-2.6.5.jar程序。

（2）創(chuàng)建HDFS數(shù)據(jù)目錄

    創(chuàng)建一個目錄，用于保存MapReduce任務(wù)的輸入文件：

[root@leaf ~]# hadoop fs -mkdir -p /data/wordcount

    創(chuàng)建一個目錄，用于保存MapReduce任務(wù)的輸出文件：

[root@leaf ~]# hadoop fs -mkdir /output

    查看剛剛創(chuàng)建的兩個目錄：

[root@leaf ~]# hadoop fs -ls /
drwxr-xr-x   - root supergroup          0 2017-09-01 20:34 /data
drwxr-xr-x   - root supergroup          0 2017-09-01 20:35 /output

（3）創(chuàng)建一個單詞文件，并上傳到HDFS

    創(chuàng)建的單詞文件如下：

[root@leaf ~]# cat myword.txt 
leaf yyh
yyh xpleaf
katy ling
yeyonghao leaf
xpleaf katy

    上傳該文件到HDFS中：

[root@leaf ~]# hadoop fs -put myword.txt /data/wordcount

    在HDFS中查看剛剛上傳的文件及內(nèi)容：

[root@leaf ~]# hadoop fs -ls /data/wordcount
-rw-r--r--   1 root supergroup         57 2017-09-01 20:40 /data/wordcount/myword.txt
[root@leaf ~]# hadoop fs -cat /data/wordcount/myword.txt
leaf yyh
yyh xpleaf
katy ling
yeyonghao leaf
xpleaf katy

（4）運行wordcount程序

    執(zhí)行如下命令：

[root@leaf ~]# hadoop jar /usr/local/hadoop/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.6.5.jar wordcount /data/wordcount /output/wordcount
...
17/09/01 20:48:14 INFO mapreduce.Job: Job job_local1719603087_0001 completed successfully
17/09/01 20:48:14 INFO mapreduce.Job: Counters: 38
        File System Counters
                FILE: Number of bytes read=585940
                FILE: Number of bytes written=1099502
                FILE: Number of read operations=0
                FILE: Number of large read operations=0
                FILE: Number of write operations=0
                HDFS: Number of bytes read=114
                HDFS: Number of bytes written=48
                HDFS: Number of read operations=15
                HDFS: Number of large read operations=0
                HDFS: Number of write operations=4
        Map-Reduce Framework
                Map input records=5
                Map output records=10
                Map output bytes=97
                Map output materialized bytes=78
                Input split bytes=112
                Combine input records=10
                Combine output records=6
                Reduce input groups=6
                Reduce shuffle bytes=78
                Reduce input records=6
                Reduce output records=6
                Spilled Records=12
                Shuffled Maps =1
                Failed Shuffles=0
                Merged Map outputs=1
                GC time elapsed (ms)=92
                CPU time spent (ms)=0
                Physical memory (bytes) snapshot=0
                Virtual memory (bytes) snapshot=0
                Total committed heap usage (bytes)=241049600
        Shuffle Errors
                BAD_ID=0
                CONNECTION=0
                IO_ERROR=0
                WRONG_LENGTH=0
                WRONG_MAP=0
                WRONG_REDUCE=0
        File Input Format Counters 
                Bytes Read=57
        File Output Format Counters 
                Bytes Written=48

（5）查看統(tǒng)計結(jié)果

    如下：

[root@leaf ~]# hadoop fs -cat /output/wordcount/part-r-00000
katy    2
leaf    2
ling    1
xpleaf  2
yeyonghao       1
yyh     2

3.參考資料

http://www.aboutyun.com/thread-7713-1-1.html