1. MapReduce进程

一个完整的MapReduce程序在分布式运行时有三类实例进程：

MrAppMaster：负责整个程序的过程调度及状态协调。
MapTask：负责Map阶段的整个数据处理流程。
ReduceTask：负责Reduce阶段的整个数据处理流程。

2. 常用数据序列化类型

Java类型	Hadoop Writable类型
Boolean	BooleanWritable
Byte	ByteWritable
Int	IntWritable
Float	FloatWritable
Long	LongWritable
Double	DoubleWritable
String	Text
Map	MapWritable
Array	ArrayWritable
Null	NullWritable

3. MapReduce编程规范

用户编写的程序分成三个部分：Mapper、Reducer和Driver。、

1．Mapper阶段

用户自定义的Mapper要继承自己的父类
Mapper的输入数据是KV对的形式（KV的类型可自定义）
Mapper中的业务逻辑写在map()方法中
Mapper的输出数据是KV对的形式（KV的类型可自定义）
map()方法（MapTask进程）对每一个<K,V>调用一次

2．Reducer阶段

Reducer的输入数据类型对应Mapper的输出数据类型，也是KV
Reducer的业务逻辑写在reduce()方法中
ReduceTask进程对每一组相同k的<k,v>组调用一次reduce()方法

3．Driver阶段相当于YARN集群的客户端，用于提交我们整个程序到YARN集群，提交的是封装了MapReduce程序相关运行参数的job对象

4. WordCount案例实操

4.1 本地测试

需求

在给定的文本文件中统计输出每一个单词出现的总次数

输入数据
期望输出数据

atguigu 2

banzhang 1

cls 2

hadoop 1

jiao 1

ss 2

xue 1

需求分析

按照MapReduce编程规范，分别编写Mapper，Reducer，Driver。

环境准备

创建maven工程，MapReduceDemo
在pom.xml文件中添加如下依赖

<dependencies>
    <dependency>
        <groupId>org.apache.hadoop</groupId>
        <artifactId>hadoop-client</artifactId>
        <version>3.1.3</version>
    </dependency>
    <dependency>
        <groupId>junit</groupId>
        <artifactId>junit</artifactId>
        <version>4.12</version>
    </dependency>
    <dependency>
        <groupId>org.slf4j</groupId>
        <artifactId>slf4j-log4j12</artifactId>
        <version>1.7.30</version>
    </dependency>
</dependencies>

在项目的src/main/resources目录下，新建一个文件，命名为“log4j.properties”，在文件中填入。

log4j.rootLogger=INFO, stdout 
log4j.appender.stdout=org.apache.log4j.ConsoleAppender  
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout 
log4j.appender.stdout.layout.ConversionPattern=%d %p [%c] - %m%n 
log4j.appender.logfile=org.apache.log4j.FileAppender 
log4j.appender.logfile.File=target/spring.log 
log4j.appender.logfile.layout=org.apache.log4j.PatternLayout 
log4j.appender.logfile.layout.ConversionPattern=%d %p [%c] - %m%n

创建包名：com.learning.mapreduce.wordcount

编写程序

编写Mapper类

package com.panda.mapreduce.wordcount;
import java.io.IOException;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

public class WordCountMapper extends Mapper<LongWritable, Text, Text, IntWritable>{
    
    Text k = new Text();
    IntWritable v = new IntWritable(1);
  
    @Override
    protected void map(LongWritable key, Text value, Context context)    throws IOException, InterruptedException {     

        // 1 获取一行
        String line = value.toString();

        // 2 切割
        String[] words = line.split(" ");

        // 3 输出
        for (String word : words) {

            k.set(word);
            context.write(k, v);
        }
    }
}

编写Reducer类

package com.panda.mapreduce.wordcount;
import java.io.IOException;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer; 

public class WordCountReducer extends Reducer<Text, IntWritable, Text, IntWritable>{ 

    int sum;
    IntWritable v = new IntWritable(); 

    @Override
    protected void reduce(Text key, Iterable<IntWritable> values,Context context) throws IOException, InterruptedException {       

        // 1 累加求和
        sum = 0;
        for (IntWritable count : values) {
            sum += count.get();
        }
       
        // 2 输出
         v.set(sum);
        context.write(key,v);
    }
}

编写Driver驱动类

package com.panda.mapreduce.wordcount;
import java.io.IOException;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
 
public class WordCountDriver {

    public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {

        // 1 获取配置信息以及获取job对象
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf);

        // 2 关联本Driver程序的jar
        job.setJarByClass(WordCountDriver.class);

        // 3 关联Mapper和Reducer的jar
        job.setMapperClass(WordCountMapper.class);
        job.setReducerClass(WordCountReducer.class);

        // 4 设置Mapper输出的kv类型
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(IntWritable.class);

        // 5 设置最终输出kv类型
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);  

        // 6 设置输入和输出路径
        FileInputFormat.setInputPaths(job, new Path(args[0]));
        FileOutputFormat.setOutputPath(job, new Path(args[1]));

        // 7 提交job
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 0 : 1);
    }
}

本地测试

需要首先配置好HADOOP_HOME变量以及Windows运行依赖
在IDEA上运行程序

4.2. 提交到集群测试

集群上测试

用maven打jar包，需要添加的打包插件依赖

<build>
    <plugins>
        <plugin>
            <artifactId>maven-compiler-plugin</artifactId>
            <version>3.6.1</version>
            <configuration>
                <source>1.8</source>
                <target>1.8</target>
            </configuration>
        </plugin>
        <plugin>
            <artifactId>maven-assembly-plugin</artifactId>
            <configuration>
                <descriptorRefs>
                    <descriptorRef>jar-with-dependencies</descriptorRef>
                </descriptorRefs>
            </configuration>
            <executions>
                <execution>
                    <id>make-assembly</id>
                    <phase>package</phase>
                    <goals>
                        <goal>single</goal>
                    </goals>
                </execution>
            </executions>
        </plugin>
    </plugins>
</build>

将程序打成jar包
修改不带依赖的jar包名称为wc.jar，并拷贝该jar包到Hadoop集群的/opt/module/hadoop-3.1.3路径。
启动Hadoop集群

[panda@Hadoop102 hadoop-3.1.3]sbin/start-dfs.sh

[panda@Hadoop103 hadoop-3.1.3]$ sbin/start-yarn.sh

执行WordCount程序

[panda@Hadoop102 hadoop-3.1.3]$ hadoop jar  wc.jar
 com.panda.mapreduce.wordcount.WordCountDriver /user/panda/input /user/panda/output

MapReduce概述 (二)