MapReduce二次排序

MapReduce二次排序

➜  student git:(master) ✗ hadoop dfs -cat /sortTwo/data
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

18/08/05 16:18:55 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
4   3
4   2
4   1
3   4
2   7
2   3
3   1
3   2
3   3

package cn.chen.hd.mr;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.*;
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.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;
import org.apache.hadoop.mapreduce.lib.partition.HashPartitioner;

import java.io.IOException;

public class SeconadrySort{
    public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();
        Job job = new Job(conf, "SeconadrySort");

        job.setJarByClass(SeconadrySort.class);

        job.setMapperClass(SeconadryMapper.class);
        job.setReducerClass(SeconadryReducer.class);

        job.setPartitionerClass(KeyPartitioner.class);
        job.setSortComparatorClass(SortComparator.class);

        job.setInputFormatClass(TextInputFormat.class);
        job.setOutputFormatClass(TextOutputFormat.class);

        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(NullWritable.class);
        job.setOutputKeyClass(NullWritable.class);
        job.setOutputValueClass(Text.class);
        //        job.setGroupingComparatorClass(GroupingComparator.class);

        FileInputFormat.addInputPath(job, new Path("hdfs://localhost:9000/sortTwo/data"));
        FileOutputFormat.setOutputPath(job, new Path("hdfs://localhost:9000/sortTwo/result"));
        // 设置Reduce个数为1，这样才是全局排序
        job.setNumReduceTasks(1);

        System.out.println(job.waitForCompletion(true) ? 0 : 1);
    }
}

class SeconadryMapper extends Mapper<LongWritable, Text, Text, NullWritable> {
    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        context.write(value,NullWritable.get());
    }
}

class SeconadryReducer extends Reducer<Text, NullWritable, NullWritable, Text> {

    @Override
    protected void reduce(Text key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {
        for(NullWritable value : values){
            context.write(NullWritable.get(), key);
        }
    }
}

// 按照第一个排序字段进行分发
class KeyPartitioner extends HashPartitioner<Text, NullWritable> {
    @Override
    public int getPartition(Text key, NullWritable value, int numReduceTasks) {
        return (key.toString().split("\t")[0].hashCode() & Integer.MAX_VALUE) % numReduceTasks;
    }
}

class SortComparator extends WritableComparator {
    protected SortComparator() {
        super(Text.class, true);
    }

    @Override
    public int compare(WritableComparable key1, WritableComparable key2) {
        // 如果第一个排序字段相同，则比较第二个排序字段
        if(Integer.parseInt(key1.toString().split("\t")[0]) == Integer.parseInt(key2.toString().split("\t")[0])){
            if(Integer.parseInt(key1.toString().split("\t")[1]) > Integer.parseInt(key2.toString().split("\t")[1])){
                return 1;
            } else if (Integer.parseInt(key1.toString().split("\t")[1]) < Integer.parseInt(key2.toString().split("\t")[1])){
                return -1;
            } else if (Integer.parseInt(key1.toString().split("\t")[1]) == Integer.parseInt(key2.toString().split("\t")[1])){
                return 0;
            }
        } else {
            if(Integer.parseInt(key1.toString().split("\t")[0]) > Integer.parseInt(key2.toString().split("\t")[0])){
                return 1;
            } else if (Integer.parseInt(key1.toString().split("\t")[0]) < Integer.parseInt(key2.toString().split("\t")[0])){
                return -1;
            }
        }
        return 0;
    }
}

/**
 * 分组器
 */
class GroupingComparator extends WritableComparator{
    protected GroupingComparator(){
        super(Text.class, true);
    }

    @Override
    public int compare(WritableComparable w1, WritableComparable w2) {
        if(Integer.parseInt(w1.toString().split("\t")[0]) == Integer.parseInt(w2.toString().split("\t")[0]){
            return 0;
        } else if (Integer.parseInt(w1.toString().split("\t")[0]) > Integer.parseInt(w2.toString().split("\t")[0]){
            return 1;
        } else if (Integer.parseInt(w1.toString().split("\t")[0]) < Integer.parseInt(w2.toString().split("\t")[0]){
            return -1;
        }
        return 0;
    }
}

➜  student git:(master) ✗ hadoop dfs -cat /sortTwo/result/part-r-00000
DEPRECATED: Use of this script to execute hdfs command is deprecated.
Instead use the hdfs command for it.

18/08/05 16:18:28 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
2   3
2   7
3   1
3   2
3   3
3   4
4   1
4   2
4   3

在reduce函数之前按照key的前半部分进行排序，也就是第一个排序字段进行分组，main函数要加上

job.setGroupingComparatorClass(GroupingComparator.class);