MapReduce(三)：分区、排序、合并

1.分区

      实现分区的步骤：
1.1先分析一下具体的业务逻辑，确定大概有多少个分区
1.2首先书写一个类，它要继承org.apache.hadoop.mapreduce.Partitioner这个类
1.3重写public int getPartition这个方法，根据具体逻辑，读数据库或者配置返回相同的数字
1.4在main方法中设置Partioner的类，job.setPartitionerClass(DataPartitioner.class);
1.5设置Reducer的数量，job.setNumReduceTasks(6);

           以下例子是统计同一手机号的上行流量、下行流量以及总流量，要求分区

DataInfo.java

import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;

import org.apache.hadoop.io.Writable;

public class DataInfo implements Writable{
    private String tel;//手机号
    private long upFlow;//上行流量
	private long downFlow;//下行流量
    private long sumFlow;//总流量
	
    public DataInfo(){}
    public DataInfo(String tel,long upFlow,long downFlow)
    {
    	this.tel=tel;
    	this.upFlow=upFlow;
    	this.downFlow=downFlow;
    	this.sumFlow=upFlow+downFlow;
    }
    
    @Override//序列化成流
	public void write(DataOutput out) throws IOException {
		out.writeUTF(tel);
		out.writeLong(upFlow);
		out.writeLong(downFlow);
		out.writeLong(sumFlow);
	}
    @Override//反序列化成对象,注意顺序不要错了
	public void readFields(DataInput in) throws IOException {
	   	this.tel=in.readUTF();
	   	this.upFlow=in.readLong();
	   	this.downFlow=in.readLong();
	   	this.sumFlow=in.readLong();
	}
    
	@Override
	public String toString() {
		return (upFlow+"\t"+downFlow+"\t"+sumFlow);
	}
	public String getTel() {
		return tel;
	}

	public void setTel(String tel) {
		this.tel = tel;
	}

	public long getUpFlow() {
		return upFlow;
	}

	public void setUpFlow(long upFlow) {
		this.upFlow = upFlow;
	}

	public long getDownFlow() {
		return downFlow;
	}

	public void setDownFlow(long downFlow) {
		this.downFlow = downFlow;
	}

	public long getSumFlow() {
		return sumFlow;
	}

	public void setSumFlow(long sumFlow) {
		this.sumFlow = sumFlow;
	} 
}

DataCount.java

import java.io.IOException;
import java.util.HashMap;
import java.util.Map;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Partitioner;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;


public class DataCount {

	//Map
	public static class DCMapper extends Mapper<LongWritable,Text,Text,DataInfo>
	{
		private Text text=new Text();
		protected void map(LongWritable key,Text value,Context context) throws IOException, InterruptedException
		{
			String line=value.toString();
			String[] str=line.split("	");
			String tel=str[0];
			long up=Long.parseLong(str[1]);
			long down=Long.parseLong(str[2]);
			DataInfo data=new DataInfo(tel,up,down);
			text.set(tel);
			context.write(text, data);
		}
	}
	//Partition
	public static class DCPartitioner extends Partitioner<Text,DataInfo>
	{
		private static Map<String,Integer>provider=new HashMap<String,Integer>();
		static{
			provider.put("134", 1);
			provider.put("134", 1);
			provider.put("135", 2);
			provider.put("135", 2);
			provider.put("136", 3);
			provider.put("136", 3);
		}
		@Override
		public int getPartition(Text key, DataInfo data, int arg2) {
			//向数据库或配置信息读写
			String tel=key.toString().substring(0, 3);
			Integer num=provider.get(tel);
			if(num==null)
				num=0;
			return num;
		}
		
	}
	//Reducer
	public static class DCReducer extends Reducer<Text,DataInfo,Text,DataInfo>
	{
		protected void reduce(Text key,Iterable<DataInfo> values,Context context) throws IOException, InterruptedException
		{
			long up=0;
			long down=0;
			for(DataInfo data:values)
			{
				up+=data.getUpFlow();
				down+=data.getDownFlow();
			}
			DataInfo dataInfo=new DataInfo("",up,down);
			context.write(key, dataInfo);
		}
	}
	public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
         Configuration conf=new Configuration();
         Job job=Job.getInstance(conf,"patition");
         
         job.setJarByClass(DataCount.class);
         
         job.setMapperClass(DCMapper.class);
         job.setMapOutputKeyClass(Text.class);
         job.setMapOutputValueClass(DataInfo.class);
        
         job.setReducerClass(DCReducer.class);
         job.setOutputKeyClass(Text.class);
         job.setOutputValueClass(DataInfo.class);
         
         job.setPartitionerClass(DCPartitioner.class);
         
         FileInputFormat.setInputPaths(job, new Path(args[0]));
         FileOutputFormat.setOutputPath(job, new Path(args[1]));
         
         job.setNumReduceTasks(Integer.parseInt(args[2]));//设置Reduce数量，即分区数量，这里最少为3，因为分区是3
         
         System.exit(job.waitForCompletion(true)? 0:1);
	}

}

2.排序

    排序MR默认是按key2进行排序的，如果想自定义排序规则，被排序的对象要实现WritableComparable接口
   ，在compareTo方法中实现排序规则(MapReduce的shuffer会自动调用这个方法)，然后将这个对象当做k2，即可完成排序

    部分代码如下：

	@Override
	public int compareTo(InfoBean o) {
		if(this.income == o.getIncome()){
			return this.expenses > o.getExpenses() ? 1 : -1;
		}
		return this.income > o.getIncome() ? 1 : -1;
	}

3.合并

   combiner的作用就是在map端对输出先做一次合并(其实相当于一个reducer)，以减少传输到reducer的数据量。

   如果不用combiner，那么，所有的结果都是reduce完成，效率会相对低下。使用combiner，先完成的map会在本地聚合，提升速度。

注意：Combiner的输出是Reducer的输入，如果Combiner是可插拔的，添加Combiner绝不能改变最终的计算结果。所以Combiner只应该用于那种Reduce的输入key/value与输出key/value类型完全一致，且不影响最终结果的场景。比如累加，最大值等。

下面看一个排序索引的例子

a.txt: hello tom
       hello jerry
       .....
b.txt: hello jerry
       hello tom
       ....
输出：
   hello a.txt->2b.txt->2

      ......
---------------------------------
Map阶段
<0,"hello tom">
....
context.write("hello->a.txt",1);
context.write("hello->a.txt",1);
context.write("hello->a.txt",1);
context.write("hello->a.txt",1);
context.write("hello->a.txt",1);


context.write("hello->b.txt",1);
context.write("hello->b.txt",1);
context.write("hello->b.txt",1);
--------------------------------------------------------
combiner阶段
<"hello->a.txt",1>
<"hello->a.txt",1>
<"hello->a.txt",1>
<"hello->a.txt",1>
<"hello->a.txt",1>


<"hello->b.txt",1>
<"hello->b.txt",1>
<"hello->b.txt",1>


context.write("hello","a.txt->5");
context.write("hello","b.txt->3");
--------------------------------------------------------
Reducer阶段
<"hello",{"a.txt->5","b.txt->3"}>

context.write("hello","a.txt->5 b.txt->3");
-------------------------------------------------------
结果：
hello "a.txt->5 b.txt->3"
tom "a.txt->2 b.txt->1"
kitty "a.txt->1"
.......

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.mapred.FileSplit;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;


public class Combine {

	//Map
	public static class CBMapper extends Mapper<Object,Text,Text,Text>
	{
		private Text k=new Text();
		private Text v=new Text();
		protected void map(Object key,Text value,Context context) throws IOException, InterruptedException
		{
			String line=value.toString();
			String[]str=line.split("	");
			FileSplit inputSplit=(FileSplit)context.getInputSplit();
			String path=inputSplit.getPath().toString();//得到路径，通过context这个上下文得到
			for(String word:str)
			{
				k.set(word+"->"+path);
				v.set("1");
				context.write(k, v);
			}
			
		}
	}
	//Combiner(Reduce1)
	public static class CBCombiner extends Reducer<Text,Text,Text,Text>
	{
		private Text k=new Text();
		private Text v=new Text();
		protected void reduce(Text key,Iterable<Text>values,Context context) throws IOException, InterruptedException
		{
			String line=key.toString();
			String[]str=line.split("->");
			String k1=str[0];
			String path=str[1];
			int count=0;
			for(Text t:values)
			{
				count+=Integer.parseInt(t.toString());
			}
			k.set(k1);
			v.set(path+"->"+count);
			context.write(k,v);
		}
	}
	
	//Reduce(Reduce2)
	public static class CBReducer extends Reducer<Text,Text,Text,Text>
	{
		private Text v=new Text();
		@Override
		protected void reduce(Text key, Iterable<Text> values,Context context)
				throws IOException, InterruptedException {
		  String result="";
		  for(Text t:values)
		  {
			  result+=t.toString()+"\t";
		  }
		  v.set(result);
		  context.write(key, v);
		}
		
	}
	public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException 
	{
		//构建job对象
		Configuration conf=new Configuration();
		Job job=Job.getInstance(conf, "Combiner");
		
		//设置main方法所在的类
		job.setJarByClass(Combine.class);
		
		//设置mapper相关属性
		job.setMapperClass(CBMapper.class);
		job.setMapOutputKeyClass(Text.class);
		job.setMapOutputValueClass(Text.class);
		
		//设置Combiner相关属性
		
		job.setCombinerClass(CBCombiner.class);
		
		//设置reducer相关属性
		job.setReducerClass(CBReducer.class);
		job.setOutputKeyClass(Text.class);
		job.setOutputValueClass(Text.class);
		
		//设置文件输入输出
		FileInputFormat.setInputPaths(job, new Path(args[0]));
		FileOutputFormat.setOutputPath(job,new Path(args[1]));
		
		//提交任务
		System.exit(job.waitForCompletion(true)? 0:1);
	}
    
}