MapReduce实战之WordCount案例

WordCount案例

1.1 需求1:统计一堆文件中单词出现的个数

0)需求:在一堆给定的文本文件中统计输出每一个单词出现的总次数

1)数据准备:

hello world
atguigu atguigu
hadoop 
spark
hello world
atguigu atguigu
hadoop 
spark
hello world
atguigu atguigu
hadoop 
spark

2)分析

       按照mapreduce编程规范,分别编写MapperReducerDriver

 

3)编写程序

1)编写mapper

package com.atguigu.mapreduce;

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);

              }

       }

}

2)编写reducer

package com.atguigu.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>{

 

       @Override

       protected void reduce(Text key, Iterable<IntWritable> value,

                     Context context) throws IOException, InterruptedException {

             

              // 1 累加求和

              int sum = 0;

              for (IntWritable count : value) {

                     sum += count.get();

              }

             

              // 2 输出

              context.write(key, new IntWritable(sum));

       }

}

3)编写驱动类

package com.atguigu.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

              Configuration configuration = new Configuration();

              Job job = Job.getInstance(configuration);

 

              // 2 设置jar加载路径

              job.setJarByClass(WordcountDriver.class);

 

              // 3 设置mapReduce

              job.setMapperClass(WordcountMapper.class);

              job.setReducerClass(WordcountReducer.class);

 

              // 4 设置map输出

              job.setMapOutputKeyClass(Text.class);

              job.setMapOutputValueClass(IntWritable.class);

 

              // 5 设置Reduce输出

              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

              boolean result = job.waitForCompletion(true);

 

              System.exit(result ? 0 : 1);

       }

}

4)集群上测试

1)将程序打成jar包,然后拷贝到hadoop集群中。

2)启动hadoop集群

3)执行wordcount程序

[atguigu@hadoop102 software]$ hadoop jar  wc.jar com.atguigu.wordcount.WordcountDriver /user/atguigu/input /user/atguigu/output1

5)本地测试

1)在windows环境上配置HADOOP_HOME环境变量。

2)在eclipse上运行程序

3)注意:如果eclipse打印不出日志,在控制台上只显示

1.log4j:WARN No appenders could be found for logger (org.apache.hadoop.util.Shell).  

2.log4j:WARN Please initialize the log4j system properly.  

3.log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info.

需要在项目的src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 

 

1.2 需求2:把单词按照ASCII码奇偶分区(Partitioner)

0)分析

1)自定义分区

package com.atguigu.mapreduce.wordcount;

import org.apache.hadoop.io.IntWritable;

import org.apache.hadoop.io.Text;

import org.apache.hadoop.mapreduce.Partitioner;

 

public class WordCountPartitioner extends Partitioner<Text, IntWritable>{

 

       @Override

       public int getPartition(Text key, IntWritable value, int numPartitions) {

             

              // 1 获取单词key 

              String firWord = key.toString().substring(0, 1);

              char[] charArray = firWord.toCharArray();

              int result = charArray[0];

              // int result  = key.toString().charAt(0);

 

              // 2 根据奇数偶数分区

              if (result % 2 == 0) {

                     return 0;

              }else {

                     return 1;

              }

       }

}

2)在驱动中配置加载分区,设置reducetask个数

job.setPartitionerClass(WordCountPartitioner.class);

job.setNumReduceTasks(2);

 

1.3 需求3:对每一个maptask的输出局部汇总(Combiner)

0)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。

1)数据准备:

hello world
atguigu atguigu
hadoop 
spark
hello world
atguigu atguigu
hadoop 
spark
hello world
atguigu atguigu
hadoop 
spark

方案一

1)增加一个WordcountCombiner类继承Reducer

package com.atguigu.mr.combiner;

import java.io.IOException;

import org.apache.hadoop.io.IntWritable;

import org.apache.hadoop.io.Text;

import org.apache.hadoop.mapreduce.Reducer;

 

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

 

       @Override

       protected void reduce(Text key, Iterable<IntWritable> values,

                     Context context) throws IOException, InterruptedException {

        // 1 汇总

              int count = 0;

              for(IntWritable v :values){

                     count += v.get();

              }

              // 2 写出

              context.write(key, new IntWritable(count));

       }

}

2)在WordcountDriver驱动类中指定combiner

// 9 指定需要使用combiner,以及用哪个类作为combiner的逻辑

job.setCombinerClass(WordcountCombiner.class);

方案二

1)将WordcountReducer作为combiner在WordcountDriver驱动类中指定

// 指定需要使用combiner,以及用哪个类作为combiner的逻辑

job.setCombinerClass(WordcountReducer.class);

 

运行程序

 

1.4 需求4:大量小文件的切片优化(CombineTextInputFormat)

0)需求:将输入的大量小文件合并成一个切片统一处理。

1)输入数据:准备5个小文件

2)实现过程

(1)不做任何处理,运行需求1中的wordcount程序,观察切片个数为5

 

(2)在WordcountDriver中增加如下代码,运行程序,并观察运行的切片个数为1

// 如果不设置InputFormat,它默认用的是TextInputFormat.class

job.setInputFormatClass(CombineTextInputFormat.class);

CombineTextInputFormat.setMaxInputSplitSize(job, 4194304);// 4m

CombineTextInputFormat.setMinInputSplitSize(job, 2097152);// 2m

 

 

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值