hadoop编程：分析CSDN注册邮箱分布情况

标签： hadoop 编程分析 | 发表时间：2014-07-08 23:48 | 作者：jdh99

出处：http://blog.csdn.net

本文博客链接: http://blog.csdn.net/jdh99,作者:jdh,转载请注明.

环境：

主机:Ubuntu10.04

hadoop版本:1.2.1

开发工具:eclipse4.4.0

说明:

要求：原始数据共6428632条，分析不同邮箱的注册情况，并按使用人数从大到小排序。

分析：hadoop自带一个排序，是按key值来进行排序的。要按值（value）进行排序，需要二次排序。

步骤：

1.job1：统计不同注册邮箱的使用人数，用默认的key值排序，保存在HDFS系统中

2.job2：对job1的输出进行二次排序，按值从大到小排序

结果输出:

使用人数在1W以上的邮箱共有24个：

qq.com   1976196
163.com   1766927
126.com   807895
sina.com   351596
yahoo.com.cn   205491
hotmail.com   202948
gmail.com   186843
sohu.com   104736
yahoo.cn   87048
tom.com   72365
yeah.net   53295
21cn.com   50710
vip.qq.com   35119
139.com   29207
263.net   24779
sina.com.cn   19156
live.cn   18920
sina.cn   18601
yahoo.com   18454
foxmail.com   16432
163.net   15176
msn.com   14211
eyou.com   13372
yahoo.com.tw   10810

源代码:

JOB1：统计不同注册邮箱的人数

CsdnData.java

  package com.bazhangkeji.hadoop;
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 org.apache.hadoop.util.GenericOptionsParser;

public class CsdnData 
{
	public static void main(String[] args) throws Exception 
	{
		Configuration conf = new Configuration();
		String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs();
		if (otherArgs.length != 2) 
		{
			System.err.println("Usage: csdndata <in> <out>");
			System.exit(2);
		}
		Job job = new Job(conf, "csdndata");
		
		job.setJarByClass(CsdnData.class);
		job.setMapperClass(MapData.class);
		
		job.setReducerClass(ReducerData.class); 
		job.setOutputKeyClass(Text.class);
		
		job.setOutputValueClass(IntWritable.class);
		FileInputFormat.addInputPath(job, new Path(otherArgs[0]));
		FileOutputFormat.setOutputPath(job, new Path(otherArgs[1]));
		System.exit(job.waitForCompletion(true) ? 0 : 1);
  	}
}

MapData.java

package com.bazhangkeji.hadoop;
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;
import org.apache.hadoop.mapreduce.Mapper.Context;

public class MapData extends Mapper<Object, Text, Text, IntWritable>
{
	IntWritable one = new IntWritable(1);
  	Text word = new Text();

  	public void map(Object key, Text value, Context context) throws IOException, InterruptedException 
	{
  		StringBuffer str_in = new StringBuffer();
		StringBuffer str_out = new StringBuffer();
		int index = 0;
		
		//初始化字符串
		str_in.setLength(0);
		str_out.setLength(0);
		str_in.append(value.toString());
		
		//获得邮箱的起始位置
		index = str_in.toString().lastIndexOf('@');
		if (index != -1)
		{
			word.set(str_in.toString().substring(index + 1).trim().toLowerCase());
  			context.write(word, one);
		}
  	}
}

ReducerData.java

package com.bazhangkeji.hadoop;
import java.io.IOException;

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

public class ReducerData extends Reducer<Text,IntWritable,Text,IntWritable> 
{
	IntWritable result = new IntWritable();

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

JOB2：对job1的输出进行二次排序，按值从大到小排序

SortSecond.java

  package com.bazhangkeji.hadoop2;
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 org.apache.hadoop.util.GenericOptionsParser;

public class SortSecond 
{
	public static void main(String[] args) throws Exception 
	{
		Configuration conf = new Configuration();
		String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs();
		if (otherArgs.length != 2) 
		{
			System.err.println("Usage: csdndata <in> <out>");
			System.exit(2);
		}
		Job job = new Job(conf, "sortsecond");
		job.setJarByClass(SortSecond.class);
		
		job.setMapperClass(MapSecond.class);
		job.setReducerClass(ReduceSecond.class); 
		
		job.setSortComparatorClass(SortMy.class); //设置自定义二次排序策略
		
		job.setOutputKeyClass(KeyMy.class);
		job.setOutputValueClass(IntWritable.class);
		
		FileInputFormat.addInputPath(job, new Path(otherArgs[0]));
		FileOutputFormat.setOutputPath(job, new Path(otherArgs[1]));
		System.exit(job.waitForCompletion(true) ? 0 : 1);
  	}
}

MapSecond.java

  package com.bazhangkeji.hadoop2;
import java.io.IOException;
import java.util.StringTokenizer;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Mapper.Context;

public class MapSecond extends Mapper<LongWritable, Text, KeyMy, IntWritable>
{
	IntWritable one = new IntWritable(1);
  	Text word = new Text();
  	KeyMy keymy = new KeyMy();

  	public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException 
	{
  		String str_in = value.toString();
  		int index = 0;
  		
  		index = str_in.indexOf('\t');
  		if (value.toString().length() > 3 && index != -1)
  		{
	  		String str1 = str_in.substring(0, index);
	  		String str2 = str_in.substring(index + 1);
	  		
	  		if (str1.length() != 0 && str2.length() != 0)
	  		{
	  			one.set(Integer.parseInt(str2));
	  			word.set(str1);
				keymy.setFirstKey(word);
				keymy.setSecondKey(one);
	  			context.write(keymy, one);
	  		}
  		}
  	}
}

ReduceSecond.java

  package com.bazhangkeji.hadoop2;
import java.io.IOException;

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

public class ReduceSecond extends Reducer<KeyMy,IntWritable,Text,IntWritable> 
{
	IntWritable result = new IntWritable();

  	public void reduce(KeyMy key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException 
	{
    	context.write(key.getFirstKey(), key.getSecondKey());
  	}
}

KeyMy.java

  package com.bazhangkeji.hadoop2;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.WritableComparable;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
/**
 * 自定义组合键
 */
public class KeyMy implements WritableComparable<KeyMy>{
    private static final Logger logger = LoggerFactory.getLogger(KeyMy.class);
    private Text firstKey;
    private IntWritable secondKey;
    public KeyMy() {
        this.firstKey = new Text();
        this.secondKey = new IntWritable();
    }
    public Text getFirstKey() {
        return this.firstKey;
    }
    public void setFirstKey(Text firstKey) {
        this.firstKey = firstKey;
    }
    public IntWritable getSecondKey() {
        return this.secondKey;
    }
    public void setSecondKey(IntWritable secondKey) {
        this.secondKey = secondKey;
    }
    @Override
    public void readFields(DataInput dateInput) throws IOException {
        // TODO Auto-generated method stub
        this.firstKey.readFields(dateInput);
        this.secondKey.readFields(dateInput);
    }
    @Override
    public void write(DataOutput outPut) throws IOException {
        this.firstKey.write(outPut);
        this.secondKey.write(outPut);
    }
    /**
     * 自定义比较策略
     * 注意：该比较策略用于 mapreduce的第一次默认排序，也就是发生在map阶段的sort小阶段，
     * 发生地点为环形缓冲区(可以通过io.sort.mb进行大小调整)
     */
    @Override
    public int compareTo(KeyMy KeyMy) {
        logger.info("-------KeyMy flag-------");
        return this.firstKey.compareTo(KeyMy.getFirstKey());
    }
}

SortMy.java

  package com.bazhangkeji.hadoop2;

import org.apache.hadoop.io.WritableComparable;
import org.apache.hadoop.io.WritableComparator;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
/**
 * 自定义二次排序策略
 */
public class SortMy extends WritableComparator {
    private static final Logger logger = LoggerFactory.getLogger(SortMy.class);
    public SortMy() {
        super(KeyMy.class,true);
    }
    @Override
    public int compare(WritableComparable KeyMyOne,
            WritableComparable KeyMyOther) 
    {
        logger.info("---------enter SortMy flag---------");
                                                                                                                                                                                             
        KeyMy c1 = (KeyMy) KeyMyOne;
        KeyMy c2 = (KeyMy) KeyMyOther;
        
        return c2.getSecondKey().get()-c1.getSecondKey().get();//0,负数,正数
    }
}

参考资料:

1.《hadoop权威指南》

2. http://zengzhaozheng.blog.51cto.com/8219051/1379271

作者：jdh99 发表于2014/7/8 15:48:05 原文链接

阅读：64 评论：0 查看评论

Hadoop Streaming 编程

- - 学着站在巨人的肩膀上

Hadoop Streaming是Hadoop提供的一个编程工具，它允许用户使用任何可执行文件或者脚本文件作为Mapper和Reducer，例如：. 采用shell脚本语言中的一些命令作为mapper和reducer（cat作为mapper，wc作为reducer）. 本文安排如下，第二节介绍Hadoop Streaming的原理，第三节介绍Hadoop Streaming的使用方法，第四节介绍Hadoop Streaming的程序编写方法，在这一节中，用C++、C、shell脚本和python实现了WordCount作业，第五节总结了常见的问题.

hadoop编程：分析CSDN注册邮箱分布情况

相关 [hadoop 编程分析] 推荐：

Hadoop Streaming 编程