Hadoop中的MapReduce框架原理、Combiner 合并案例实操

文章目录

• 13.MapReduce框架原理
• • 13.3 Shuffle机制
  • • 13.3.9 Combiner 合并案例实操
    • • 13.3.9.1 需求
      • • 13.3.9.1.1 数据输入
        • 13.3.9.1.2 期望输出数据
      • 13.3.9.2 需求分析
      • 13.3.9.3 案例实操-方案一
      • • 13.3.9.3.1 增加一个 WordCountCombiner 类继承 Reducer
        • 13.3.9.3.2在 WordcountDriver 驱动类中指定 Combiner
      • 13.3.9.4 案例实操-方案二
      • • 13.3.9.4.1 将 WordcountReducer 作为 Combiner 在 WordcountDriver 驱动类中指定
      • 13.3.9.5 不进入Reducer

13.MapReduce框架原理

13.3 Shuffle机制

13.3.9 Combiner 合并案例实操

13.3.9.1 需求

  统计过程中对每一个 MapTask 的输出进行局部汇总，以减小网络传输量即采用Combiner 功能。

13.3.9.1.1 数据输入

13.3.9.1.2 期望输出数据

  期望：Combine 输入数据多，输出时经过合并，输出数据降低。

13.3.9.2 需求分析

创建一个combiner的文件夹，将wordcount里面3个java代码同时复制到combiner里面

13.3.9.3 案例实操-方案一

13.3.9.3.1 增加一个 WordCountCombiner 类继承 Reducer

package com.summer.mapreduce.combiner;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-10-04 17:21
 */
public class WordCountCombiner extends Reducer<Text, IntWritable, Text, IntWritable> {

    private IntWritable outV = new IntWritable();
    @Override
    protected void reduce(Text key, Iterable<IntWritable> values, Reducer<Text, IntWritable, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        int sum = 0;
        for (IntWritable value : values) {
            sum += value.get();
        }
        outV.set(sum);

        context.write(key, outV);
    }
}

13.3.9.3.2在 WordcountDriver 驱动类中指定 Combiner

// 指定需要使用combiner，以及用哪个类作为combiner的逻辑
job.setCombinerClass(WordCountCombiner.class);

package com.summer.mapreduce.combiner;

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;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-08-22 17:23
 */

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

        //1 获取job
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf);

        //2 设置jar包路径
        job.setJarByClass(WordCountDriver.class);

        //3 关联mapper和reduccer
        job.setMapperClass(WordCountMapper.class);
        job.setReducerClass(WordCountReducer.class);

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

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

        // 指定需要使用combiner，以及用哪个类作为combiner的逻辑
        job.setCombinerClass(WordCountCombiner.class);

        //6 设置输入路径和输出路径
        FileInputFormat.setInputPaths(job, new Path("D:\\Acode\\Hadoop\\input\\inputhello"));
        //输出的路径为空，要是有该文件，则会报错
        FileOutputFormat.setOutputPath(job, new Path("D:\\Acode\\Hadoop\\output\\output1"));

        //7 提交job
        boolean result = job.waitForCompletion(true);

        System.exit(result ? 0 : 1);
    }
}

使用combiner的input和output和materialized的大小

不使用combiner的input和output和materialized的大小

13.3.9.4 案例实操-方案二

package com.summer.mapreduce.combiner;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-10-04 17:21
 */
public class WordCountCombiner extends Reducer<Text, IntWritable, Text, IntWritable> {

    private IntWritable outV = new IntWritable();
    @Override
    protected void reduce(Text key, Iterable<IntWritable> values, Reducer<Text, IntWritable, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        int sum = 0;
        for (IntWritable value : values) {
            sum += value.get();
        }
        outV.set(sum);

        context.write(key, outV);
    }
}

package com.summer.mapreduce.combiner;


import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-08-22 17:23
 */


/**
 * KEYIN, reduce阶段输入的key的类型：Text
 * VALUEIN, reduce阶段输入的value的类型：IntWritable
 * KEYOUT, reduce阶段输出的key的类型：Text
 * VALUEOUT,reduce阶段输出的kvalue的类型：IntWritable
 */
public class WordCountReducer extends Reducer<Text, IntWritable,Text,IntWritable> {
    private IntWritable outV = new IntWritable();
    @Override
    protected void reduce(Text key, Iterable<IntWritable> values, Reducer<Text, IntWritable, Text, IntWritable>.Context context) throws IOException, InterruptedException {

        int sum = 0;
        //ha(1,1)
        //累加
        for (IntWritable value : values) {
            sum += value.get();
        }

        outV.set(sum);

        //写出
        context.write(key, outV);

    }
}

因为自定义的Combiner和Reducer的代码是一样的，所以可以调用Reducer作为Combiner

13.3.9.4.1 将 WordcountReducer 作为 Combiner 在 WordcountDriver 驱动类中指定

package com.summer.mapreduce.combiner;

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;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-08-22 17:23
 */

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

        //1 获取job
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf);

        //2 设置jar包路径
        job.setJarByClass(WordCountDriver.class);

        //3 关联mapper和reduccer
        job.setMapperClass(WordCountMapper.class);
        job.setReducerClass(WordCountReducer.class);

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

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

        // 指定需要使用combiner，以及用哪个类作为combiner的逻辑
//        job.setCombinerClass(WordCountCombiner.class);
        // 不进入Reducer
//        job.setNumReduceTasks(0);

        //使用Reducer，而不使用自定义的combiner
        job.setCombinerClass(WordCountReducer.class);

        //6 设置输入路径和输出路径
        FileInputFormat.setInputPaths(job, new Path("D:\\Acode\\Hadoop\\input\\inputhello"));
        //输出的路径为空，要是有该文件，则会报错
        FileOutputFormat.setOutputPath(job, new Path("D:\\Acode\\Hadoop\\output\\output5"));

        //7 提交job
        boolean result = job.waitForCompletion(true);

        System.exit(result ? 0 : 1);
    }
}

结果和前面几种情况是一致的，over！

13.3.9.5 不进入Reducer

package com.summer.mapreduce.combiner;

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;

import java.io.IOException;

/**
 * @author Redamancy
 * @create 2022-08-22 17:23
 */

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

        //1 获取job
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf);

        //2 设置jar包路径
        job.setJarByClass(WordCountDriver.class);

        //3 关联mapper和reduccer
        job.setMapperClass(WordCountMapper.class);
        job.setReducerClass(WordCountReducer.class);

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

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

        // 指定需要使用combiner，以及用哪个类作为combiner的逻辑
        job.setCombinerClass(WordCountCombiner.class);
        // 不进入Reducer
        job.setNumReduceTasks(0);

        //6 设置输入路径和输出路径
        FileInputFormat.setInputPaths(job, new Path("D:\\Acode\\Hadoop\\input\\inputhello"));
        //输出的路径为空，要是有该文件，则会报错
        FileOutputFormat.setOutputPath(job, new Path("D:\\Acode\\Hadoop\\output\\output4"));

        //7 提交job
        boolean result = job.waitForCompletion(true);

        System.exit(result ? 0 : 1);
    }
}

这个结果是带m，而使用Reducer的时候是带r