Hadoop入门-WordCount示例

WordCount的过程如图，这里记录下入门的过程，虽然有很多地方理解的只是皮毛。

Hadoop的安装

安装比较简单，安装完成后进行单机环境的配置。

hadoop-env.sh:指定JAVA_HOME。

# The only required environment variable is JAVA_HOME.  All others are
# optional.  When running a distributed configuration it is best to
# set JAVA_HOME in this file, so that it is correctly defined on
# remote nodes.

# The java implementation to use.
export JAVA_HOME="$(/usr/libexec/java_home)"

core-site.xml：设置Hadoop使用的临时目录，NameNode的地址。

<configuration>
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/usr/local/Cellar/hadoop/hdfs/tmp</value>
    </property>
    <property>
        <name>fs.default.name</name>
        <value>hdfs://localhost:9000</value>
    </property>
</configuration>

hdfs-site.xml：一个节点，副本个数设为1。

<configuration>
    <property>
        <name>dfs.replication</name>
        <value>1</value>
    </property>
</configuration>

mapred-site.xml:指定JobTracker的地址。

<configuration>
    <property>
        <name>mapred.job.tracker</name>
        <value>localhost:9010</value>
    </property>
</configuration>

启动Hadoop相关的所有进程。

➜  sbin git:(master) ./start-all.sh
This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh
16/12/03 19:32:18 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Starting namenodes on [localhost]
Password:
localhost: starting namenode, logging to /usr/local/Cellar/hadoop/2.7.1/libexec/logs/hadoop-vonzhou-namenode-vonzhoudeMacBook-Pro.local.out
Password:
localhost: starting datanode, logging to /usr/local/Cellar/hadoop/2.7.1/libexec/logs/hadoop-vonzhou-datanode-vonzhoudeMacBook-Pro.local.out
Starting secondary namenodes [0.0.0.0]
Password:
0.0.0.0: starting secondarynamenode, logging to /usr/local/Cellar/hadoop/2.7.1/libexec/logs/hadoop-vonzhou-secondarynamenode-vonzhoudeMacBook-Pro.local.out
16/12/03 19:33:27 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
starting yarn daemons
starting resourcemanager, logging to /usr/local/Cellar/hadoop/2.7.1/libexec/logs/yarn-vonzhou-resourcemanager-vonzhoudeMacBook-Pro.local.out
Password:
localhost: starting nodemanager, logging to /usr/local/Cellar/hadoop/2.7.1/libexec/logs/yarn-vonzhou-nodemanager-vonzhoudeMacBook-Pro.local.out

（可以配置ssh无密码登录方式，否则启动hadoop的时候总是要密码。）

看看启动了哪些组件。

➜  sbin git:(master) jps -l
5713 org.apache.hadoop.hdfs.server.namenode.NameNode
6145 org.apache.hadoop.yarn.server.nodemanager.NodeManager
6044 org.apache.hadoop.yarn.server.resourcemanager.ResourceManager
5806 org.apache.hadoop.hdfs.server.datanode.DataNode
5918 org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode

访问http://localhost:50070/可以看到DFS的一些状态。

WordCount 单词计数

WordCount就是Hadoop学习的hello world，代码如下：

public class WordCount {

    public static class Map extends Mapper<LongWritable, Text, Text, IntWritable> {
        private final static IntWritable one = new IntWritable(1);
        private Text word = new Text();

        public void map(LongWritable key, Text value, Context context)
                throws IOException, InterruptedException {
            String line = value.toString();
            StringTokenizer tokenizer = new StringTokenizer(line);
            while (tokenizer.hasMoreTokens()) {
                word.set(tokenizer.nextToken());
                context.write(word, one);
            }
        }
    }

    public static class Reduce extends
            Reducer<Text, IntWritable, Text, IntWritable> {

        public void reduce(Text key, Iterable<IntWritable> values,
                           Context context) throws IOException, InterruptedException {
            int sum = 0;
            for (IntWritable val : values) {
                sum += val.get();
            }
            context.write(key, new IntWritable(sum));
        }
    }

    public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();

        Job job = new Job(conf, "wordcount");
        job.setJarByClass(WordCount.class);

        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);

        job.setMapperClass(Map.class);
        job.setReducerClass(Reduce.class);
        /**
         * 设置一个本地combine,可以极大的消除本节点重复单词的计数,减小网络传输的开销
         */
        job.setCombinerClass(Reduce.class);

        job.setInputFormatClass(TextInputFormat.class);
        job.setOutputFormatClass(TextOutputFormat.class);

        FileInputFormat.addInputPath(job, new Path(args[0]));
        FileOutputFormat.setOutputPath(job, new Path(args[1]));

        job.waitForCompletion(true);
    }
}

构造两个文本文件, 把本地的两个文件拷贝到HDFS中：

➜  hadoop-examples git:(master) ✗ ln /usr/local/Cellar/hadoop/2.7.1/bin/hadoop hadoop
➜  hadoop-examples git:(master) ✗ ./hadoop dfs -put wordcount-input/file* input
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

16/12/03 23:17:10 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
➜  hadoop-examples git:(master) ✗ ./hadoop dfs -ls input/      
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

16/12/03 23:21:08 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Found 2 items
-rw-r--r--   1 vonzhou supergroup         42 2016-12-03 23:17 input/file1
-rw-r--r--   1 vonzhou supergroup         43 2016-12-03 23:17 input/file2

编译程序得到jar:

mvn clean package

运行程序（指定main class的时候需要全包名限定）：

➜  hadoop-examples git:(master) ✗ ./hadoop jar target/hadoop-examples-1.0-SNAPSHOT.jar com.vonzhou.learnhadoop.simple.WordCount input output
16/12/03 23:31:19 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
16/12/03 23:31:20 INFO Configuration.deprecation: session.id is deprecated. Instead, use dfs.metrics.session-id
16/12/03 23:31:20 INFO jvm.JvmMetrics: Initializing JVM Metrics with processName=JobTracker, sessionId=
16/12/03 23:33:21 WARN mapreduce.JobResourceUploader: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
16/12/03 23:33:21 INFO input.FileInputFormat: Total input paths to process : 2
16/12/03 23:33:21 INFO mapreduce.JobSubmitter: number of splits:2
16/12/03 23:33:22 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_local524341653_0001
16/12/03 23:33:22 INFO mapreduce.Job: The url to track the job: http://localhost:8080/
16/12/03 23:33:22 INFO mapreduce.Job: Running job: job_local524341653_0001
16/12/03 23:33:22 INFO mapred.LocalJobRunner: OutputCommitter set in config null
16/12/03 23:33:22 INFO output.FileOutputCommitter: File Output Committer Algorithm version is 1
16/12/03 23:33:22 INFO mapred.LocalJobRunner: OutputCommitter is org.apache.hadoop.mapreduce.lib.output.FileOutputCommitter
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Waiting for map tasks
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Starting task: attempt_local524341653_0001_m_000000_0
16/12/03 23:33:22 INFO output.FileOutputCommitter: File Output Committer Algorithm version is 1
16/12/03 23:33:22 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
16/12/03 23:33:22 INFO mapred.Task:  Using ResourceCalculatorProcessTree : null
16/12/03 23:33:22 INFO mapred.MapTask: Processing split: hdfs://localhost:9000/user/vonzhou/input/file2:0+43
16/12/03 23:33:22 INFO mapred.MapTask: (EQUATOR) 0 kvi 26214396(104857584)
16/12/03 23:33:22 INFO mapred.MapTask: mapreduce.task.io.sort.mb: 100
16/12/03 23:33:22 INFO mapred.MapTask: soft limit at 83886080
16/12/03 23:33:22 INFO mapred.MapTask: bufstart = 0; bufvoid = 104857600
16/12/03 23:33:22 INFO mapred.MapTask: kvstart = 26214396; length = 6553600
16/12/03 23:33:22 INFO mapred.MapTask: Map output collector class = org.apache.hadoop.mapred.MapTask$MapOutputBuffer
16/12/03 23:33:22 INFO mapred.LocalJobRunner: 
16/12/03 23:33:22 INFO mapred.MapTask: Starting flush of map output
16/12/03 23:33:22 INFO mapred.MapTask: Spilling map output
16/12/03 23:33:22 INFO mapred.MapTask: bufstart = 0; bufend = 71; bufvoid = 104857600
16/12/03 23:33:22 INFO mapred.MapTask: kvstart = 26214396(104857584); kvend = 26214372(104857488); length = 25/6553600
16/12/03 23:33:22 INFO mapred.MapTask: Finished spill 0
16/12/03 23:33:22 INFO mapred.Task: Task:attempt_local524341653_0001_m_000000_0 is done. And is in the process of committing
16/12/03 23:33:22 INFO mapred.LocalJobRunner: map
16/12/03 23:33:22 INFO mapred.Task: Task 'attempt_local524341653_0001_m_000000_0' done.
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Finishing task: attempt_local524341653_0001_m_000000_0
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Starting task: attempt_local524341653_0001_m_000001_0
16/12/03 23:33:22 INFO output.FileOutputCommitter: File Output Committer Algorithm version is 1
16/12/03 23:33:22 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
16/12/03 23:33:22 INFO mapred.Task:  Using ResourceCalculatorProcessTree : null
16/12/03 23:33:22 INFO mapred.MapTask: Processing split: hdfs://localhost:9000/user/vonzhou/input/file1:0+42
16/12/03 23:33:22 INFO mapred.MapTask: (EQUATOR) 0 kvi 26214396(104857584)
16/12/03 23:33:22 INFO mapred.MapTask: mapreduce.task.io.sort.mb: 100
16/12/03 23:33:22 INFO mapred.MapTask: soft limit at 83886080
16/12/03 23:33:22 INFO mapred.MapTask: bufstart = 0; bufvoid = 104857600
16/12/03 23:33:22 INFO mapred.MapTask: kvstart = 26214396; length = 6553600
16/12/03 23:33:22 INFO mapred.MapTask: Map output collector class = org.apache.hadoop.mapred.MapTask$MapOutputBuffer
16/12/03 23:33:22 INFO mapred.LocalJobRunner: 
16/12/03 23:33:22 INFO mapred.MapTask: Starting flush of map output
16/12/03 23:33:22 INFO mapred.MapTask: Spilling map output
16/12/03 23:33:22 INFO mapred.MapTask: bufstart = 0; bufend = 70; bufvoid = 104857600
16/12/03 23:33:22 INFO mapred.MapTask: kvstart = 26214396(104857584); kvend = 26214372(104857488); length = 25/6553600
16/12/03 23:33:22 INFO mapred.MapTask: Finished spill 0
16/12/03 23:33:22 INFO mapred.Task: Task:attempt_local524341653_0001_m_000001_0 is done. And is in the process of committing
16/12/03 23:33:22 INFO mapred.LocalJobRunner: map
16/12/03 23:33:22 INFO mapred.Task: Task 'attempt_local524341653_0001_m_000001_0' done.
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Finishing task: attempt_local524341653_0001_m_000001_0
16/12/03 23:33:22 INFO mapred.LocalJobRunner: map task executor complete.
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Waiting for reduce tasks
16/12/03 23:33:22 INFO mapred.LocalJobRunner: Starting task: attempt_local524341653_0001_r_000000_0
16/12/03 23:33:22 INFO output.FileOutputCommitter: File Output Committer Algorithm version is 1
16/12/03 23:33:22 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
16/12/03 23:33:22 INFO mapred.Task:  Using ResourceCalculatorProcessTree : null
16/12/03 23:33:22 INFO mapred.ReduceTask: Using ShuffleConsumerPlugin: org.apache.hadoop.mapreduce.task.reduce.Shuffle@64accbd9
16/12/03 23:33:23 INFO mapreduce.Job: Job job_local524341653_0001 running in uber mode : false
16/12/03 23:33:23 INFO mapreduce.Job:  map 100% reduce 0%
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: MergerManager: memoryLimit=334338464, maxSingleShuffleLimit=83584616, mergeThreshold=220663392, ioSortFactor=10, memToMemMergeOutputsThreshold=10
16/12/03 23:33:53 INFO reduce.EventFetcher: attempt_local524341653_0001_r_000000_0 Thread started: EventFetcher for fetching Map Completion Events
16/12/03 23:33:53 INFO reduce.LocalFetcher: localfetcher#1 about to shuffle output of map attempt_local524341653_0001_m_000001_0 decomp: 86 len: 90 to MEMORY
16/12/03 23:33:53 INFO reduce.InMemoryMapOutput: Read 86 bytes from map-output for attempt_local524341653_0001_m_000001_0
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: closeInMemoryFile -> map-output of size: 86, inMemoryMapOutputs.size() -> 1, commitMemory -> 0, usedMemory ->86
16/12/03 23:33:53 INFO reduce.LocalFetcher: localfetcher#1 about to shuffle output of map attempt_local524341653_0001_m_000000_0 decomp: 87 len: 91 to MEMORY
16/12/03 23:33:53 INFO reduce.InMemoryMapOutput: Read 87 bytes from map-output for attempt_local524341653_0001_m_000000_0
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: closeInMemoryFile -> map-output of size: 87, inMemoryMapOutputs.size() -> 2, commitMemory -> 86, usedMemory ->173
16/12/03 23:33:53 INFO reduce.EventFetcher: EventFetcher is interrupted.. Returning
16/12/03 23:33:53 INFO mapred.LocalJobRunner: 2 / 2 copied.
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: finalMerge called with 2 in-memory map-outputs and 0 on-disk map-outputs
16/12/03 23:33:53 INFO mapred.Merger: Merging 2 sorted segments
16/12/03 23:33:53 INFO mapred.Merger: Down to the last merge-pass, with 2 segments left of total size: 162 bytes
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: Merged 2 segments, 173 bytes to disk to satisfy reduce memory limit
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: Merging 1 files, 175 bytes from disk
16/12/03 23:33:53 INFO reduce.MergeManagerImpl: Merging 0 segments, 0 bytes from memory into reduce
16/12/03 23:33:53 INFO mapred.Merger: Merging 1 sorted segments
16/12/03 23:33:53 INFO mapred.Merger: Down to the last merge-pass, with 1 segments left of total size: 165 bytes
16/12/03 23:33:53 INFO mapred.LocalJobRunner: 2 / 2 copied.
16/12/03 23:33:53 INFO Configuration.deprecation: mapred.skip.on is deprecated. Instead, use mapreduce.job.skiprecords
16/12/03 23:33:53 INFO mapred.Task: Task:attempt_local524341653_0001_r_000000_0 is done. And is in the process of committing
16/12/03 23:33:53 INFO mapred.LocalJobRunner: 2 / 2 copied.
16/12/03 23:33:53 INFO mapred.Task: Task attempt_local524341653_0001_r_000000_0 is allowed to commit now
16/12/03 23:33:53 INFO output.FileOutputCommitter: Saved output of task 'attempt_local524341653_0001_r_000000_0' to hdfs://localhost:9000/user/vonzhou/output/_temporary/0/task_local524341653_0001_r_000000
16/12/03 23:33:53 INFO mapred.LocalJobRunner: reduce > reduce
16/12/03 23:33:53 INFO mapred.Task: Task 'attempt_local524341653_0001_r_000000_0' done.
16/12/03 23:33:53 INFO mapred.LocalJobRunner: Finishing task: attempt_local524341653_0001_r_000000_0
16/12/03 23:33:53 INFO mapred.LocalJobRunner: reduce task executor complete.
16/12/03 23:33:54 INFO mapreduce.Job:  map 100% reduce 100%
16/12/03 23:33:54 INFO mapreduce.Job: Job job_local524341653_0001 completed successfully
16/12/03 23:33:54 INFO mapreduce.Job: Counters: 35
        File System Counters
                FILE: Number of bytes read=54188
                FILE: Number of bytes written=917564
                FILE: Number of read operations=0
                FILE: Number of large read operations=0
                FILE: Number of write operations=0
                HDFS: Number of bytes read=213
                HDFS: Number of bytes written=89
                HDFS: Number of read operations=22
                HDFS: Number of large read operations=0
                HDFS: Number of write operations=5
        Map-Reduce Framework
                Map input records=5
                Map output records=14
                Map output bytes=141
                Map output materialized bytes=181
                Input split bytes=222
                Combine input records=0
                Combine output records=0
                Reduce input groups=11
                Reduce shuffle bytes=181
                Reduce input records=14
                Reduce output records=11
                Spilled Records=28
                Shuffled Maps =2
                Failed Shuffles=0
                Merged Map outputs=2
                GC time elapsed (ms)=7
                Total committed heap usage (bytes)=946864128
        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=85
        File Output Format Counters 
                Bytes Written=89
➜  hadoop-examples git:(master) ✗

查看执行的结果：

➜  hadoop-examples git:(master) ✗ ./hadoop dfs -ls output
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

16/12/03 23:36:42 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Found 2 items
-rw-r--r--   1 vonzhou supergroup          0 2016-12-03 23:33 output/_SUCCESS
-rw-r--r--   1 vonzhou supergroup         89 2016-12-03 23:33 output/part-r-00000
➜  hadoop-examples git:(master) ✗ ./hadoop dfs -cat output/part-r-00000
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

16/12/03 23:37:03 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
big     1
by      1
data    1
google  1
hadoop  2
hello   2
learning        1
papers  1
step    2
vonzhou 1
world   1