hadoop参数

zqdjd7g9  于 2021-06-02  发布在  Hadoop
关注(0)|答案(0)|浏览(229)

我有两份工作。第一个作业map/reduce只获取两个值:context.write(one,two)。但是在第二个jobs map/reduce中,我需要一个文件和两个值。
我试着把第一份工作的结果放在上下文中
在减速器中

@Override
     protected void cleanup(Context context)
       throws IOException, InterruptedException {
          Configuration conf = context.getConfiguration();
          conf.setFloat("one",valorOne);
          conf.setFloat("two",valorTwo);
    }

在第二个任务的Map绘制者那里得到:

@Override
      protected void setup(Context context) throws IOException,
        InterruptedException {
         Configuration conf = context.getConfiguration();
          one=conf.getFloat("one",0f);
          two=conf.getFloat("two",0f);

      }

但是值是空的。
在逃

public int run(String[] args) throws Exception {
 Job job = Job.getInstance(getConf(), "1job");
        job.setJarByClass(job.class);

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

        job.setMapOutputKeyClass(IntWritable.class);
        job.setMapOutputValueClass(ValoresWritable.class);

        job.setOutputKeyClass(IntWritable.class);
        job.setOutputValueClass(ValoresWritable.class);

        job.setMapperClass(Job1Map.class);
        job.setCombinerClass(Job1Reducer.class);
        job.setReducerClass(Job1Reducer.class);         

        FileInputFormat.addInputPath(job, new Path(input));
        FileOutputFormat.setOutputPath(job, new Path(output));

        job.waitForCompletion(true);

        Job job2 =  Job.getInstance(getConf(), "job2");
        job2.setJarByClass(job.class);

        job2.setInputFormatClass(TextInputFormat.class);
        job2.setOutputFormatClass(TextOutputFormat.class);

        job2.setMapOutputKeyClass(IntWritable.class);
        job2.setMapOutputValueClass(IntWritable.class);

        job2.setOutputKeyClass(IntWritable.class);
        job2.setOutputValueClass(IntWritable.class);

        job2.setMapperClass(Job2Map.class);
        job2.setReducerClass(Job2Reduce.class);

我做错了什么?谢谢

暂无答案!

目前还没有任何答案,快来回答吧!

相关问题