一、简介
求TopK是算法中最常使用到的，现在使用Mapreduce在海量数据中统计数据的求TopK。
二、例子
（1）实例描述
给出三个文件，每个文件中都存储了若干个数值，求所有数值中的求Top 5。
样例输入：
1）file1：
1
2
3
7
9
-99
2
2）file2：
11
2
23
17
9
199
22
3）file3：
21
12
3
17
2
39
12
期望输出：
199
39
23
22
21
（2）问题分析
实现统计海量数据的求TopK，不能将所有的数据加载到内存，计算只能使用类似外部排序的方式，加载一部分数据统计求TopK，接着加载另一部分进行统计TopK。
（3）实现步骤
1）Map过程
首先使用默认的TextInputFormat类对输入文件进行处理，得到文本中每行的偏移量及其内容。显然，Map过程首先必须分析输入的<key,value>对，得到数值，然后在mapper中统计单个分块的求TopK。
2）Reduce过程
经过map方法处理后，Reduce过程将获取每个mapper的求TopK进行统计，分行统计出总的TopK。
（3）关键代码
package com.mk.mapreduce;import org.apache.commons.lang.StringUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
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;import java.io.IOException;
import java.net.URI;
import java.util.ArrayList;
import java.util.Collections;
import java.util.List;public class TopK {public static class TopKMapper extends Mapper<LongWritable, Text, IntWritable, NullWritable> {private List<Integer> top5 = new ArrayList<>(5);@Overrideprotected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {if (StringUtils.isBlank(value.toString())) {System.out.println("空白行");return;}Integer v = Integer.valueOf(value.toString().trim());if(top5.size()<5){top5.add(v);}else{Integer min = Collections.min(top5);if (min < v) {top5.remove(min);top5.add(v);}}}@Overrideprotected void cleanup(Context context) throws IOException, InterruptedException {for (Integer v : top5)context.write(new IntWritable(v), NullWritable.get());}}public static class TopKReducer extends Reducer< IntWritable, NullWritable,IntWritable, NullWritable> {private List<Integer> top5 = new ArrayList<>(5);@Overrideprotected void reduce(IntWritable key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {Integer v = key.get();if(top5.size()<5){top5.add(v);}else{Integer min = Collections.min(top5);if (min < v) {top5.remove(min);top5.add(v);}}}@Overrideprotected void cleanup(Context context) throws IOException, InterruptedException {top5.sort((a,b)->b-a);for (Integer v : top5)context.write(new IntWritable(v), NullWritable.get());}}public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {String uri = "hdfs://192.168.150.128:9000";String input = "/topk/input";String output = "/topk/output";Configuration conf = new Configuration();if (System.getProperty("os.name").toLowerCase().contains("win"))conf.set("mapreduce.app-submission.cross-platform", "true");FileSystem fileSystem = FileSystem.get(URI.create(uri), conf);Path path = new Path(output);fileSystem.delete(path, true);Job job = new Job(conf, "TopK");job.setJar("./out/artifacts/hadoop_test_jar/hadoop-test.jar");job.setJarByClass(TopK.class);job.setMapperClass(TopKMapper.class);job.setReducerClass(TopKReducer.class);job.setMapOutputKeyClass(IntWritable.class);job.setMapOutputValueClass(NullWritable.class);job.setOutputKeyClass(IntWritable.class);job.setOutputValueClass(NullWritable.class);FileInputFormat.addInputPaths(job, uri + input);FileOutputFormat.setOutputPath(job, new Path(uri + output));boolean ret = job.waitForCompletion(true);System.out.println(job.getJobName() + "-----" + ret);}
}
