Run Hadoop job without using JobConf

Question

I can't find a single example of submitting a Hadoop job that does not use the deprecated JobConf class. JobClient, which hasn't been deprecated, still only supports methods that take a JobConf parameter.

Can someone please point me at an example of Java code submitting a Hadoop map/reduce job using only the Configuration class (not JobConf), and using the mapreduce.lib.input package instead of mapred.input?

score 23 · Accepted Answer · answered Jan 22 '10 at 09:12

Hope this helpful

import java.io.File;

import org.apache.commons.io.FileUtils;
import org.apache.hadoop.conf.Configured;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.util.Tool;
import org.apache.hadoop.util.ToolRunner;

public class MapReduceExample extends Configured implements Tool {

    static class MyMapper extends Mapper<LongWritable, Text, LongWritable, Text> {
        public MyMapper(){

        }

        protected void map(
                LongWritable key,
                Text value,
                org.apache.hadoop.mapreduce.Mapper<LongWritable, Text, LongWritable, Text>.Context context)
                throws java.io.IOException, InterruptedException {
            context.getCounter("mygroup", "jeff").increment(1);
            context.write(key, value);
        };
    }

    @Override
    public int run(String[] args) throws Exception {
        Job job = new Job();
        job.setMapperClass(MyMapper.class);
        FileInputFormat.setInputPaths(job, new Path(args[0]));
        FileOutputFormat.setOutputPath(job, new Path(args[1]));

        job.waitForCompletion(true);
        return 0;
    }

    public static void main(String[] args) throws Exception {
        FileUtils.deleteDirectory(new File("data/output"));
        args = new String[] { "data/input", "data/output" };
        ToolRunner.run(new MapReduceExample(), args);
    }
}

All three `Job` constructors are deprecated now. The correct way is: `Job job = Job.getInstance(getConf());` — Majid Azimi, Jan 06 '13 at 07:54
On which version? I am using v1.0.4 but not found this constructor. — Abhendra Singh, Feb 19 '13 at 11:55

score 9 · Answer 2 · answered Jan 23 '10 at 20:20

9

I believe this tutorial illustrates removing the deprecated JobConf class using Hadoop 0.20.1.

answered Jan 23 '10 at 20:20

Binary Nerd

13,872
4
42
44

score 2 · Answer 3 · answered Apr 20 '12 at 17:15

2

This is a nice example with downloadable code: http://sonerbalkir.blogspot.com/2010/01/new-hadoop-api-020x.html It's also over two years old and there is no official documentation discussing the new API. Sad.

answered Apr 20 '12 at 17:15

dk.

2,030
1
22
22

score 1 · Answer 4 · answered Apr 08 '13 at 04:35

In the previous API there were three ways of submitting the job and one of them is by submitting the job and getting a reference to the RunningJob and getting an id of the RunningJob.

submitJob(JobConf) : only submits the job, then poll the returned handle to the RunningJob to query status and make scheduling decisions.

How can one use the new Api and get a reference to the RunningJob and get an id of the runningJob as none of the api's return a reference to RunningJob

http://hadoop.apache.org/docs/current/api/org/apache/hadoop/mapreduce/Job.html

thanks

coderz · Answer 5 · 2015-03-31T09:51:05.503

Try to use Configuration and Job. Here is an example:

(Replace your Mapper, Combiner, Reducer classes and other configuration)

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

public class WordCount {
  public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
    Configuration conf = new Configuration();
    if(args.length != 2) {
      System.err.println("Usage: <in> <out>");
      System.exit(2);
    }
    Job job = Job.getInstance(conf, "Word Count");

    // set jar
    job.setJarByClass(WordCount.class);

    // set Mapper, Combiner, Reducer
    job.setMapperClass(TokenizerMapper.class);
    job.setCombinerClass(IntSumReducer.class);
    job.setReducerClass(IntSumReducer.class);

    /* Optional, set customer defined Partioner:
     * job.setPartitionerClass(MyPartioner.class);
     */

    // set output key
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(IntWritable.class);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(IntWritable.class);

    // set input and output path
    FileInputFormat.addInputPath(job, new Path(args[0]));
    FileOutputFormat.setOutputPath(job, new Path(args[1]));

    // by default, Hadoop use TextInputFormat and TextOutputFormat
    // any customer defined input and output class must implement InputFormat/OutputFormat interface
    job.setInputFormatClass(TextInputFormat.class);
    job.setOutputFormatClass(TextOutputFormat.class);

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

Run Hadoop job without using JobConf

5 Answers5

Linked