MapReduce -- Hello World

通常我们在学习一门语言的时候，写的第一个程序就是Hello World。而在学习Hadoop时，我们要写的第一个程序就是词频统计WordCount程序，这是一个官方的例子。这个例子中涉及到了HDFS,MapReduce，YARN，如果它能正常跑完，也就证明了hadoop 的基础组件运行没有问题。

这里面的核心组件是MapReduce，它由两个词组成，承担的任务是：

• Map 任务分解
• Reduce 结果汇总

运行WordCount程序

运行环境：HDP 2.6

切换用户，避免路径和权限问题：

$ sudo su - hdfs

1. 生成测试文件 test.txt

随便在里面写文字（最后统计这些文字出现次数）

2. 将文件上传 hdfs

$ hadoop fs -mkdir /input
$ hadoop fs -put ./test.txt /input/test.txt

3. 查找示例文件的路径

$ find /usr/hdp -name "hadoop-mapreduce-example*"
/usr/hdp/2.6.4.0-91/hadoop-mapreduce/hadoop-mapreduce-examples.jar
/usr/hdp/2.6.4.0-91/hadoop-mapreduce/hadoop-mapreduce-examples-2.7.3.2.6.4.0-91.jar

4. 运行hadoop实例

$ hadoop jar /usr/hdp/2.6.4.0-91/hadoop-mapreduce/hadoop-mapreduce-examples.jar wordcount /input /out
18/10/15 07:46:18 INFO client.AHSProxy: Connecting to Application History server at my-vm-hdp-1/192.168.0.68:10200
18/10/15 07:46:18 INFO client.RequestHedgingRMFailoverProxyProvider: Looking for the active RM in [rm1, rm2]...
18/10/15 07:46:18 INFO client.RequestHedgingRMFailoverProxyProvider: Found active RM [rm2]
18/10/15 07:46:18 INFO input.FileInputFormat: Total input paths to process : 1
18/10/15 07:46:18 INFO mapreduce.JobSubmitter: number of splits:1
18/10/15 07:46:18 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1539582562189_0004
18/10/15 07:46:19 INFO impl.YarnClientImpl: Submitted application application_1539582562189_0004
18/10/15 07:46:19 INFO mapreduce.Job: The url to track the job: [http://my-vm-hdp-2:8088/proxy/application_1539582562189_0004/](http://my-vm-hdp-2:8088/proxy/application_1539582562189_0004/)
18/10/15 07:46:19 INFO mapreduce.Job: Running job: job_1539582562189_0004
18/10/15 07:46:24 INFO mapreduce.Job: Job job_1539582562189_0004 running in uber mode : false
18/10/15 07:46:24 INFO mapreduce.Job:  map 0% reduce 0%
18/10/15 07:46:28 INFO mapreduce.Job:  map 100% reduce 0%
18/10/15 07:46:33 INFO mapreduce.Job:  map 100% reduce 100%
18/10/15 07:46:33 INFO mapreduce.Job: Job job_1539582562189_0004 completed successfully
18/10/15 07:46:33 INFO mapreduce.Job: Counters: 49
    File System Counters
        FILE: Number of bytes read=2578
        FILE: Number of bytes written=310487
        FILE: Number of read operations=0
        FILE: Number of large read operations=0
        FILE: Number of write operations=0
        HDFS: Number of bytes read=3636
        HDFS: Number of bytes written=1933
        HDFS: Number of read operations=6
        HDFS: Number of large read operations=0
        HDFS: Number of write operations=2
    Job Counters
        Launched map tasks=1
        Launched reduce tasks=1
        Data-local map tasks=1
        Total time spent by all maps in occupied slots (ms)=10048
        Total time spent by all reduces in occupied slots (ms)=4134
        Total time spent by all map tasks (ms)=2512
        Total time spent by all reduce tasks (ms)=2067
        Total vcore-milliseconds taken by all map tasks=2512
        Total vcore-milliseconds taken by all reduce tasks=2067
        Total megabyte-milliseconds taken by all map tasks=5144576
        Total megabyte-milliseconds taken by all reduce tasks=2116608
    Map-Reduce Framework
        Map input records=59
        Map output records=383
        Map output bytes=4821
        Map output materialized bytes=2578
        Input split bytes=97
        Combine input records=383
        Combine output records=161
        Reduce input groups=161
        Reduce shuffle bytes=2578
        Reduce input records=161
        Reduce output records=161
        Spilled Records=322
        Shuffled Maps =1
        Failed Shuffles=0
        Merged Map outputs=1
        GC time elapsed (ms)=88
        CPU time spent (ms)=1850
        Physical memory (bytes) snapshot=1617342464
        Virtual memory (bytes) snapshot=6485979136
        Total committed heap usage (bytes)=1510998016
    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=3539
    File Output Format Counters
        Bytes Written=1933

另一个例子，计算圆周率pi值：

$ hadoop jar /usr/hdp/2.6.4.0-91/hadoop-mapreduce/hadoop-mapreduce-examples.jar pi 5 5

更多的参数参考这里。

5. 运行过程中，打开另一个窗口，yarn命令检查application

$ yarn application -list
18/10/15 07:54:13 INFO client.AHSProxy: Connecting to Application History server at my-vm-hdp-1/192.168.0.68:10200
18/10/15 07:54:13 INFO client.RequestHedgingRMFailoverProxyProvider: Looking for the active RM in [rm1, rm2]...
18/10/15 07:54:13 INFO client.RequestHedgingRMFailoverProxyProvider: Found active RM [rm2]
Total number of applications (application-types: [] and states: [SUBMITTED, ACCEPTED, RUNNING]):1
                Application-Id        Application-Name        Application-Type          User         Queue                 State           Final-State           Progress                           Tracking-URL
application_1539582562189_0005              word count               MAPREDUCE          hdfs       default               RUNNING             UNDEFINED                50%    [http://my-vm-hdp-3:43645](http://my-vm-hdp-3:43645/)

6. 运行结束后，查看生成的文件

$ hadoop fs -ls /out
Found 2 items
-rw-r--r--   3 hdfs hadoop          0 2018-10-15 07:46 /out/_SUCCESS
-rw-r--r--   3 hdfs hadoop       1933 2018-10-15 07:46 /out/part-r-00000

$ hadoop fs -cat /out/part-r-00000