Examples with Job org.apache.hadoop.mapreduce.Job used on opensource projects

Example 76 with Job

use of org.apache.hadoop.mapreduce.Job in project hadoop by apache.

the class Join method run.

/**
   * The main driver for sort program.
   * Invoke this method to submit the map/reduce job.
   * @throws IOException When there is communication problems with the 
   *                     job tracker.
   */
@SuppressWarnings("unchecked")
public int run(String[] args) throws Exception {
    Configuration conf = getConf();
    JobClient client = new JobClient(conf);
    ClusterStatus cluster = client.getClusterStatus();
    int num_reduces = (int) (cluster.getMaxReduceTasks() * 0.9);
    String join_reduces = conf.get(REDUCES_PER_HOST);
    if (join_reduces != null) {
        num_reduces = cluster.getTaskTrackers() * Integer.parseInt(join_reduces);
    }
    Job job = Job.getInstance(conf);
    job.setJobName("join");
    job.setJarByClass(Sort.class);
    job.setMapperClass(Mapper.class);
    job.setReducerClass(Reducer.class);
    Class<? extends InputFormat> inputFormatClass = SequenceFileInputFormat.class;
    Class<? extends OutputFormat> outputFormatClass = SequenceFileOutputFormat.class;
    Class<? extends WritableComparable> outputKeyClass = BytesWritable.class;
    Class<? extends Writable> outputValueClass = TupleWritable.class;
    String op = "inner";
    List<String> otherArgs = new ArrayList<String>();
    for (int i = 0; i < args.length; ++i) {
        try {
            if ("-r".equals(args[i])) {
                num_reduces = Integer.parseInt(args[++i]);
            } else if ("-inFormat".equals(args[i])) {
                inputFormatClass = Class.forName(args[++i]).asSubclass(InputFormat.class);
            } else if ("-outFormat".equals(args[i])) {
                outputFormatClass = Class.forName(args[++i]).asSubclass(OutputFormat.class);
            } else if ("-outKey".equals(args[i])) {
                outputKeyClass = Class.forName(args[++i]).asSubclass(WritableComparable.class);
            } else if ("-outValue".equals(args[i])) {
                outputValueClass = Class.forName(args[++i]).asSubclass(Writable.class);
            } else if ("-joinOp".equals(args[i])) {
                op = args[++i];
            } else {
                otherArgs.add(args[i]);
            }
        } catch (NumberFormatException except) {
            System.out.println("ERROR: Integer expected instead of " + args[i]);
            return printUsage();
        } catch (ArrayIndexOutOfBoundsException except) {
            System.out.println("ERROR: Required parameter missing from " + args[i - 1]);
            // exits
            return printUsage();
        }
    }
    // Set user-supplied (possibly default) job configs
    job.setNumReduceTasks(num_reduces);
    if (otherArgs.size() < 2) {
        System.out.println("ERROR: Wrong number of parameters: ");
        return printUsage();
    }
    FileOutputFormat.setOutputPath(job, new Path(otherArgs.remove(otherArgs.size() - 1)));
    List<Path> plist = new ArrayList<Path>(otherArgs.size());
    for (String s : otherArgs) {
        plist.add(new Path(s));
    }
    job.setInputFormatClass(CompositeInputFormat.class);
    job.getConfiguration().set(CompositeInputFormat.JOIN_EXPR, CompositeInputFormat.compose(op, inputFormatClass, plist.toArray(new Path[0])));
    job.setOutputFormatClass(outputFormatClass);
    job.setOutputKeyClass(outputKeyClass);
    job.setOutputValueClass(outputValueClass);
    Date startTime = new Date();
    System.out.println("Job started: " + startTime);
    int ret = job.waitForCompletion(true) ? 0 : 1;
    Date end_time = new Date();
    System.out.println("Job ended: " + end_time);
    System.out.println("The job took " + (end_time.getTime() - startTime.getTime()) / 1000 + " seconds.");
    return ret;
}

Example 77 with Job

use of org.apache.hadoop.mapreduce.Job in project hadoop by apache.

the class CompressMapper method getCompressJob.

public static Job getCompressJob(String jobname, Configuration conf, String inputpath, String outputpath) throws Exception {
    Job job = Job.getInstance(conf, jobname + "-CompressMapperJob");
    job.setJarByClass(CompressMapper.class);
    job.setMapperClass(TextCompressMapper.class);
    job.setOutputKeyClass(Text.class);
    job.setMapOutputValueClass(Text.class);
    // if output file exists ,delete it
    final FileSystem hdfs = FileSystem.get(new ScenarioConfiguration());
    if (hdfs.exists(new Path(outputpath))) {
        hdfs.delete(new Path(outputpath), true);
    }
    hdfs.close();
    job.setInputFormatClass(SequenceFileInputFormat.class);
    FileInputFormat.addInputPath(job, new Path(inputpath));
    FileOutputFormat.setOutputPath(job, new Path(outputpath));
    return job;
}

Example 78 with Job

use of org.apache.hadoop.mapreduce.Job in project hadoop by apache.

the class TestCompressionEmulationUtils method runDataGenJob.

/**
   * Runs a GridMix data-generation job.
   */
private static void runDataGenJob(Configuration conf, Path tempDir) throws IOException, ClassNotFoundException, InterruptedException {
    JobClient client = new JobClient(conf);
    // get the local job runner
    conf.setInt(MRJobConfig.NUM_MAPS, 1);
    Job job = Job.getInstance(conf);
    CompressionEmulationUtil.configure(job);
    job.setInputFormatClass(CustomInputFormat.class);
    // set the output path
    FileOutputFormat.setOutputPath(job, tempDir);
    // submit and wait for completion
    job.submit();
    int ret = job.waitForCompletion(true) ? 0 : 1;
    assertEquals("Job Failed", 0, ret);
}

Example 79 with Job

use of org.apache.hadoop.mapreduce.Job in project hadoop by apache.

the class TestDistCacheEmulation method testGenerateDistCacheData.

/**
   * Validate GenerateDistCacheData job if it creates dist cache files properly.
   * 
   * @throws Exception
   */
@Test(timeout = 200000)
public void testGenerateDistCacheData() throws Exception {
    long[] sortedFileSizes = new long[5];
    Configuration jobConf = runSetupGenerateDistCacheData(true, sortedFileSizes);
    GridmixJob gridmixJob = new GenerateDistCacheData(jobConf);
    Job job = gridmixJob.call();
    assertEquals("Number of reduce tasks in GenerateDistCacheData is not 0.", 0, job.getNumReduceTasks());
    assertTrue("GenerateDistCacheData job failed.", job.waitForCompletion(false));
    validateDistCacheData(jobConf, sortedFileSizes);
}

Example 80 with Job

use of org.apache.hadoop.mapreduce.Job in project hadoop by apache.

the class ValueAggregatorJob method main.

/**
   * create and run an Aggregate based map/reduce job.
   * 
   * @param args the arguments used for job creation
   * @throws IOException
   */
public static void main(String[] args) throws IOException, InterruptedException, ClassNotFoundException {
    Job job = ValueAggregatorJob.createValueAggregatorJob(new Configuration(), args);
    int ret = job.waitForCompletion(true) ? 0 : 1;
    System.exit(ret);
}