MapReduce既是编程模型又是计算框架

// 计算单词数量的 MapReduce 版本
public class WordCount {

  public static class TokenizerMapper
       extends Mapper<Object, Text, Text, IntWritable>{

    private final static IntWritable one = new IntWritable(1);
    private Text word = new Text();

// map 函数 
    public void map(Object key, Text value, Context context
                    ) throws IOException, InterruptedException {
      StringTokenizer itr = new StringTokenizer(value.toString());
      while (itr.hasMoreTokens()) {
        word.set(itr.nextToken());
        context.write(word, one);
        // 对每个单词输出一个 < word, 1 > 的键值对
      }
    }
  }
  // MR 计算框架会将这些 < word, 1 > 收集起来
  // 将相同的 word 放在一起，形成 <word , <1,1,1,1,1,1,1…>> 这样的 <Key, Value 集合 > 数据
  // 然后将其输入给 reduce 函数

  public static class IntSumReducer
       extends Reducer<Text,IntWritable,Text,IntWritable> {
    private IntWritable result = new IntWritable();
	
	// reduce 函数
    public void reduce(Text key, Iterable<IntWritable> values,
                       Context context
                       ) throws IOException, InterruptedException {
      int sum = 0;
      for (IntWritable val : values) {
        sum += val.get();
        // 对 word 对应的 1 的集合 求和 sum
      }
      result.set(sum);
      context.write(key, result);
      // 输出 < word, sum > 键值对
    }
  }
}