纯代码视角看分布式运算（上（wordcount）

一、wordcount：

 1、WordCountMap      

String line = value.toString();  

            

StringTokenizer token = new StringTokenizer(line);  

       

     

while (token.hasMoreTokens()) {

 

word.set(token.nextToken());  

context.write(word, one);

  

            }  

 

2、WordCountReduce

 

 

for (IntWritable val : values) {

  

 sum += val.get();  

            }  

            

context.write(key, new IntWritable(sum));  

 

 

 

3、main

 

job.setOutputKeyClass(Text.class);  

        

job.setOutputValueClass(IntWritable.class);  

  

        

job.setMapperClass(WordCountMap.class);  

       

job.setReducerClass(WordCountReduce.class);  

  

        

job.setInputFormatClass(TextInputFormat.class);  

        

job.setOutputFormatClass(TextOutputFormat.class);  

  

        

FileInputFormat.addInputPath(job, new Path(args[0]));  

       

 FileOutputFormat.setOutputPath(job, new Path(args[1]));  

  

        

job.waitForCompletion(true);  

 

二、Wc

1、Map

 

String line = value.toString();

      StringTokenizer tokenizer = new StringTokenizer(line);

      while (tokenizer.hasMoreTokens()) {

          word.set(tokenizer.nextToken());

          output.collect(word, one);

}

 

 

2、Reduce

 

int sum = 0;

      while (values.hasNext()) {

          sum += values.next().get();

      }

      output.collect(key, new IntWritable(sum));

 

 

 

 

3、Main

 

JobConf conf = new JobConf(wc.class);

  conf.setJobName("wordcount");

 

  conf.setOutputKeyClass(Text.class);

  conf.setOutputValueClass(IntWritable.class);

 

  conf.setMapperClass(Map.class);

  conf.setCombinerClass(Reduce.class);

  conf.setReducerClass(Reduce.class);

 

  conf.setInputFormat(TextInputFormat.class);

  conf.setOutputFormat(TextOutputFormat.class);

 

  FileInputFormat.setInputPaths(conf, new Path(args[0]));

  FileOutputFormat.setOutputPath(conf, new Path(args[1]));

 

  JobClient.runJob(conf);