Example usage for org.apache.hadoop.mapred JobConf setNumReduceTasks

List of usage examples for org.apache.hadoop.mapred JobConf setNumReduceTasks

  1. HOME
  2. Java
  3. org.apache.hadoop
  4. org.apache.hadoop.mapred.*
  5. JobConf
  6. setNumReduceTasks

Introduction

In this page you can find the example usage for org.apache.hadoop.mapred JobConf setNumReduceTasks.

Prototype

public void setNumReduceTasks(int n)

Source Link

Document

Set the requisite number of reduce tasks for this job.

Usage

From source file:Brush.BrushConfig.java

License:Apache License

public static void initializeConfiguration(JobConf conf) {
    validateConfiguration();//from  w w  w.j  av  a2  s  . c om

    conf.setNumMapTasks(HADOOP_MAPPERS);
    conf.setNumReduceTasks(HADOOP_REDUCERS);
    conf.set("mapred.child.java.opts", HADOOP_JAVAOPTS);
    conf.set("mapred.task.timeout", Long.toString(HADOOP_TIMEOUT));
    conf.setLong("LOCALNODES", HADOOP_LOCALNODES);

    conf.setLong("UP_KMER", UP_KMER);
    conf.setLong("LOW_KMER", LOW_KMER);
    conf.setLong("K", K);
    //conf.setFloat("ERRORRATE", ERRORRATE);
    conf.setFloat("MAJORITY", MAJORITY);
    conf.setFloat("PWM_N", PWM_N);
    conf.setFloat("EXPCOV", EXPCOV);
    conf.setFloat("KMERCOV", KMERCOV);
    conf.setLong("READLENGTH", READLEN);
    conf.setLong("TIPLENGTH", TIPLENGTH);
    conf.setLong("INSLENGTH", INSLEN);
    conf.setLong("INSLENGTH_SD", INSLEN_SD);
    conf.setLong("MAXBUBBLELEN", MAXBUBBLELEN);
    conf.setFloat("BUBBLEEDITRATE", BUBBLEEDITRATE);
    conf.setFloat("LOW_COV_THRESH", LOW_COV_THRESH);
    conf.setLong("MAX_LOW_COV_LEN", MAX_LOW_COV_LEN);
    //conf.setFloat("ERRORRATE", ERRORRATE);

    conf.setLong("N50_TARGET", N50_TARGET);
}

From source file:Brush.Graph2Fasta.java

License:Apache License

public RunningJob run(String inputPath, String outputPath) throws Exception {
    sLogger.info("Tool name: Graph2Fasta");
    sLogger.info(" - input: " + inputPath);
    sLogger.info(" - output: " + outputPath);

    JobConf conf = new JobConf(Graph2Fasta.class);
    conf.setJobName("Graph2Fasta " + inputPath);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(Graph2FastaMapper.class);

    BrushConfig.initializeConfiguration(conf);
    conf.setNumReduceTasks(0);

    //delete the output directory if it exists already
    FileSystem.get(conf).delete(new Path(outputPath), true);

    return JobClient.runJob(conf);
}

From source file:Brush.Graph2Sfa.java

License:Apache License

public RunningJob run(String inputPath, String outputPath) throws Exception {
    sLogger.info("Tool name: Graph2Sfa");
    sLogger.info(" - input: " + inputPath);
    sLogger.info(" - output: " + outputPath);

    JobConf conf = new JobConf(Graph2Sfa.class);
    conf.setJobName("Graph2Sfa " + inputPath);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(Graph2SfaMapper.class);

    BrushConfig.initializeConfiguration(conf);
    conf.setNumReduceTasks(0);

    //delete the output directory if it exists already
    FileSystem.get(conf).delete(new Path(outputPath), true);

    return JobClient.runJob(conf);
}

From source file:Brush.Stats.java

License:Apache License

public RunningJob run(String inputPath, String outputPath) throws Exception {
    sLogger.info("Tool name: Stats");
    sLogger.info(" - input: " + inputPath);
    sLogger.info(" - output: " + outputPath);

    JobConf conf = new JobConf(Stats.class);
    conf.setJobName("Stats " + inputPath);

    BrushConfig.initializeConfiguration(conf);
    conf.setNumReduceTasks(1);

    FileInputFormat.addInputPath(conf, new Path(inputPath));
    FileOutputFormat.setOutputPath(conf, new Path(outputPath));

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(Text.class);

    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(Text.class);

    conf.setMapperClass(StatsMapper.class);
    conf.setReducerClass(StatsReducer.class);

    //delete the output directory if it exists already
    FileSystem.get(conf).delete(new Path(outputPath), true);

    return JobClient.runJob(conf);
}

From source file:ca.etsmtl.lasi.hbasewikipedialoader.HBaseWikipediaLoader.java

License:Apache License

/**
 * Sets up the actual job./*w  w  w .  j a v  a 2s.c  o m*/
 * 
 * @param conf
 *          The current configuration.
 * @param args
 *          The command line parameters.
 * @return The newly created job.
 * @throws IOException
 *           When setting up the job fails.
 */
public static JobConf createSubmittableJob(HBaseConfiguration conf, String[] args) throws IOException {
    JobConf jobConf = new JobConf(conf, HBaseWikipediaLoader.class);
    jobConf.setJobName(NAME);

    // Stream stuff
    jobConf.set("stream.recordreader.class", "org.apache.hadoop.streaming.StreamXmlRecordReader");
    jobConf.set("stream.recordreader.begin", "<page>");
    jobConf.set("stream.recordreader.end", "</page>");

    jobConf.setSpeculativeExecution(false);

    jobConf.setMapOutputKeyClass(ImmutableBytesWritable.class);
    jobConf.setMapOutputValueClass(BatchUpdate.class);

    jobConf.setMapperClass(Map.class);

    jobConf.setNumReduceTasks(0);

    jobConf.setInputFormat(StreamInputFormat.class);
    jobConf.setOutputFormat(TableOutputFormat.class);
    jobConf.set(TableOutputFormat.OUTPUT_TABLE, TABLE);
    jobConf.setOutputKeyClass(ImmutableBytesWritable.class);
    jobConf.setOutputValueClass(BatchUpdate.class);

    StreamInputFormat.setInputPaths(jobConf, new Path(args[0]));
    FileOutputFormat.setOutputPath(jobConf, new Path("/tmp/" + NAME + "-" + System.currentTimeMillis()));

    return jobConf;

}

From source file:cascading.flow.FlowStep.java

License:Open Source License

protected JobConf getJobConf(JobConf parentConf) throws IOException {
    JobConf conf = parentConf == null ? new JobConf() : new JobConf(parentConf);

    // set values first so they can't break things downstream
    if (hasProperties()) {
        for (Map.Entry entry : getProperties().entrySet())
            conf.set(entry.getKey().toString(), entry.getValue().toString());
    }//from w ww . j ava 2s .c o m

    // disable warning
    conf.setBoolean("mapred.used.genericoptionsparser", true);

    conf.setJobName(getStepName());

    conf.setOutputKeyClass(Tuple.class);
    conf.setOutputValueClass(Tuple.class);

    conf.setMapperClass(FlowMapper.class);
    conf.setReducerClass(FlowReducer.class);

    // set for use by the shuffling phase
    TupleSerialization.setSerializations(conf);

    initFromSources(conf);

    initFromSink(conf);

    initFromTraps(conf);

    if (sink.getScheme().getNumSinkParts() != 0) {
        // if no reducer, set num map tasks to control parts
        if (getGroup() != null)
            conf.setNumReduceTasks(sink.getScheme().getNumSinkParts());
        else
            conf.setNumMapTasks(sink.getScheme().getNumSinkParts());
    }

    conf.setOutputKeyComparatorClass(TupleComparator.class);

    if (getGroup() == null) {
        conf.setNumReduceTasks(0); // disable reducers
    } else {
        // must set map output defaults when performing a reduce
        conf.setMapOutputKeyClass(Tuple.class);
        conf.setMapOutputValueClass(Tuple.class);

        // handles the case the groupby sort should be reversed
        if (getGroup().isSortReversed())
            conf.setOutputKeyComparatorClass(ReverseTupleComparator.class);

        addComparators(conf, "cascading.group.comparator", getGroup().getGroupingSelectors());

        if (getGroup().isGroupBy())
            addComparators(conf, "cascading.sort.comparator", getGroup().getSortingSelectors());

        if (!getGroup().isGroupBy()) {
            conf.setPartitionerClass(CoGroupingPartitioner.class);
            conf.setMapOutputKeyClass(IndexTuple.class); // allows groups to be sorted by index
            conf.setMapOutputValueClass(IndexTuple.class);
            conf.setOutputKeyComparatorClass(IndexTupleCoGroupingComparator.class); // sorts by group, then by index
            conf.setOutputValueGroupingComparator(CoGroupingComparator.class);
        }

        if (getGroup().isSorted()) {
            conf.setPartitionerClass(GroupingPartitioner.class);
            conf.setMapOutputKeyClass(TuplePair.class);

            if (getGroup().isSortReversed())
                conf.setOutputKeyComparatorClass(ReverseGroupingSortingComparator.class);
            else
                conf.setOutputKeyComparatorClass(GroupingSortingComparator.class);

            // no need to supply a reverse comparator, only equality is checked
            conf.setOutputValueGroupingComparator(GroupingComparator.class);
        }
    }

    // perform last so init above will pass to tasks
    conf.setInt("cascading.flow.step.id", id);
    conf.set("cascading.flow.step", Util.serializeBase64(this));

    return conf;
}

From source file:cascading.flow.hadoop.HadoopFlowStep.java

License:Open Source License

public JobConf createInitializedConfig(FlowProcess<JobConf> flowProcess, JobConf parentConfig) {
    JobConf conf = parentConfig == null ? new JobConf() : HadoopUtil.copyJobConf(parentConfig);

    // disable warning
    conf.setBoolean("mapred.used.genericoptionsparser", true);

    conf.setJobName(getStepDisplayName(conf.getInt("cascading.display.id.truncate", Util.ID_LENGTH)));

    conf.setOutputKeyClass(Tuple.class);
    conf.setOutputValueClass(Tuple.class);

    conf.setMapRunnerClass(FlowMapper.class);
    conf.setReducerClass(FlowReducer.class);

    // set for use by the shuffling phase
    TupleSerialization.setSerializations(conf);

    initFromSources(flowProcess, conf);/*from   w w  w.  j a va  2s . c o  m*/

    initFromSink(flowProcess, conf);

    initFromTraps(flowProcess, conf);

    initFromStepConfigDef(conf);

    int numSinkParts = getSink().getScheme().getNumSinkParts();

    if (numSinkParts != 0) {
        // if no reducer, set num map tasks to control parts
        if (getGroup() != null)
            conf.setNumReduceTasks(numSinkParts);
        else
            conf.setNumMapTasks(numSinkParts);
    } else if (getGroup() != null) {
        int gatherPartitions = conf.getNumReduceTasks();

        if (gatherPartitions == 0)
            gatherPartitions = conf.getInt(FlowRuntimeProps.GATHER_PARTITIONS, 0);

        if (gatherPartitions == 0)
            throw new FlowException(getName(),
                    "a default number of gather partitions must be set, see FlowRuntimeProps");

        conf.setNumReduceTasks(gatherPartitions);
    }

    conf.setOutputKeyComparatorClass(TupleComparator.class);

    if (getGroup() == null) {
        conf.setNumReduceTasks(0); // disable reducers
    } else {
        // must set map output defaults when performing a reduce
        conf.setMapOutputKeyClass(Tuple.class);
        conf.setMapOutputValueClass(Tuple.class);
        conf.setPartitionerClass(GroupingPartitioner.class);

        // handles the case the groupby sort should be reversed
        if (getGroup().isSortReversed())
            conf.setOutputKeyComparatorClass(ReverseTupleComparator.class);

        addComparators(conf, "cascading.group.comparator", getGroup().getKeySelectors(), this, getGroup());

        if (getGroup().isGroupBy())
            addComparators(conf, "cascading.sort.comparator", getGroup().getSortingSelectors(), this,
                    getGroup());

        if (!getGroup().isGroupBy()) {
            conf.setPartitionerClass(CoGroupingPartitioner.class);
            conf.setMapOutputKeyClass(IndexTuple.class); // allows groups to be sorted by index
            conf.setMapOutputValueClass(IndexTuple.class);
            conf.setOutputKeyComparatorClass(IndexTupleCoGroupingComparator.class); // sorts by group, then by index
            conf.setOutputValueGroupingComparator(CoGroupingComparator.class);
        }

        if (getGroup().isSorted()) {
            conf.setPartitionerClass(GroupingSortingPartitioner.class);
            conf.setMapOutputKeyClass(TuplePair.class);

            if (getGroup().isSortReversed())
                conf.setOutputKeyComparatorClass(ReverseGroupingSortingComparator.class);
            else
                conf.setOutputKeyComparatorClass(GroupingSortingComparator.class);

            // no need to supply a reverse comparator, only equality is checked
            conf.setOutputValueGroupingComparator(GroupingComparator.class);
        }
    }

    // perform last so init above will pass to tasks
    String versionString = Version.getRelease();

    if (versionString != null)
        conf.set("cascading.version", versionString);

    conf.set(CASCADING_FLOW_STEP_ID, getID());
    conf.set("cascading.flow.step.num", Integer.toString(getOrdinal()));

    HadoopUtil.setIsInflow(conf);

    Iterator<FlowNode> iterator = getFlowNodeGraph().getTopologicalIterator();

    String mapState = pack(iterator.next(), conf);
    String reduceState = pack(iterator.hasNext() ? iterator.next() : null, conf);

    // hadoop 20.2 doesn't like dist cache when using local mode
    int maxSize = Short.MAX_VALUE;

    int length = mapState.length() + reduceState.length();

    if (isHadoopLocalMode(conf) || length < maxSize) // seems safe
    {
        conf.set("cascading.flow.step.node.map", mapState);

        if (!Util.isEmpty(reduceState))
            conf.set("cascading.flow.step.node.reduce", reduceState);
    } else {
        conf.set("cascading.flow.step.node.map.path",
                HadoopMRUtil.writeStateToDistCache(conf, getID(), "map", mapState));

        if (!Util.isEmpty(reduceState))
            conf.set("cascading.flow.step.node.reduce.path",
                    HadoopMRUtil.writeStateToDistCache(conf, getID(), "reduce", reduceState));
    }

    return conf;
}

From source file:cn.edu.xmu.dm.mapreduce.Sort.java

License:Apache License

/**
 * The main driver for sort program. Invoke this method to submit the
 * map/reduce job./*from w ww. j a  v a 2 s.com*/
 * 
 * @throws IOException
 *             When there is communication problems with the job tracker.
 */
public int run(String[] args) throws Exception {
    Configuration conf = new Configuration();
    Job job = new Job(conf, "Sorter");
    job.setJarByClass(Sort.class);

    JobConf jobConf = new JobConf(getConf(), Sort.class);
    jobConf.setJobName("sorter");

    jobConf.setMapperClass(IdentityMapper.class);
    jobConf.setReducerClass(IdentityReducer.class);

    JobClient client = new JobClient(jobConf);
    ClusterStatus cluster = client.getClusterStatus();
    int num_reduces = (int) (cluster.getMaxReduceTasks() * 0.9);
    String sort_reduces = jobConf.get("test.sort.reduces_per_host");
    if (sort_reduces != null) {
        num_reduces = cluster.getTaskTrackers() * Integer.parseInt(sort_reduces);
    }
    Class<? extends InputFormat> inputFormatClass = SequenceFileInputFormat.class;
    Class<? extends OutputFormat> outputFormatClass = SequenceFileOutputFormat.class;
    Class<? extends WritableComparable> outputKeyClass = BytesWritable.class;
    Class<? extends Writable> outputValueClass = BytesWritable.class;
    List<String> otherArgs = new ArrayList<String>();
    InputSampler.Sampler<K, V> sampler = null;
    for (int i = 0; i < args.length; ++i) {
        try {
            if ("-m".equals(args[i])) {
                jobConf.setNumMapTasks(Integer.parseInt(args[++i]));
            } else 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 ("-totalOrder".equals(args[i])) {
                double pcnt = Double.parseDouble(args[++i]);
                int numSamples = Integer.parseInt(args[++i]);
                int maxSplits = Integer.parseInt(args[++i]);
                if (0 >= maxSplits)
                    maxSplits = Integer.MAX_VALUE;
                sampler = new InputSampler.RandomSampler<K, V>(pcnt, numSamples, maxSplits);
            } 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]);
            return printUsage(); // exits
        }
    }

    // Set user-supplied (possibly default) job configs
    jobConf.setNumReduceTasks(num_reduces);

    jobConf.setInputFormat(inputFormatClass);
    jobConf.setOutputFormat(outputFormatClass);

    jobConf.setOutputKeyClass(outputKeyClass);
    jobConf.setOutputValueClass(outputValueClass);

    // Make sure there are exactly 2 parameters left.
    if (otherArgs.size() != 2) {
        System.out.println("ERROR: Wrong number of parameters: " + otherArgs.size() + " instead of 2.");
        return printUsage();
    }
    FileInputFormat.setInputPaths(jobConf, otherArgs.get(0));
    FileOutputFormat.setOutputPath(jobConf, new Path(otherArgs.get(1)));

    if (sampler != null) {
        System.out.println("Sampling input to effect total-order sort...");
        jobConf.setPartitionerClass(TotalOrderPartitioner.class);
        Path inputDir = FileInputFormat.getInputPaths(jobConf)[0];
        inputDir = inputDir.makeQualified(inputDir.getFileSystem(jobConf));
        Path partitionFile = new Path(inputDir, "_sortPartitioning");
        TotalOrderPartitioner.setPartitionFile(jobConf, partitionFile);
        InputSampler.<K, V>writePartitionFile(jobConf, sampler);
        URI partitionUri = new URI(partitionFile.toString() + "#" + "_sortPartitioning");
        DistributedCache.addCacheFile(partitionUri, jobConf);
        DistributedCache.createSymlink(jobConf);
    }

    System.out.println("Running on " + cluster.getTaskTrackers() + " nodes to sort from "
            + FileInputFormat.getInputPaths(jobConf)[0] + " into " + FileOutputFormat.getOutputPath(jobConf)
            + " with " + num_reduces + " reduces.");
    Date startTime = new Date();
    System.out.println("Job started: " + startTime);
    jobResult = JobClient.runJob(jobConf);
    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 0;
}

From source file:co.nubetech.hiho.job.DBQueryInputJob.java

License:Apache License

public void runJobs(Configuration conf, int jobCounter) throws IOException {

    try {//from   w  w w  .ja  v  a  2  s  .com
        checkMandatoryConfs(conf);
    } catch (HIHOException e1) {
        e1.printStackTrace();
        throw new IOException(e1);
    }

    Job job = new Job(conf);
    for (Entry<String, String> entry : conf) {
        logger.warn("key, value " + entry.getKey() + "=" + entry.getValue());
    }

    // logger.debug("Number of maps " +
    // conf.getInt("mapred.map.tasks", 1));
    // conf.setInt(JobContext.NUM_MAPS,
    // conf.getInt("mapreduce.job.maps", 1));
    // job.getConfiguration().setInt("mapred.map.tasks", 4);
    job.getConfiguration().setInt(MRJobConfig.NUM_MAPS, conf.getInt(HIHOConf.NUMBER_MAPPERS, 1));
    logger.warn("Number of maps " + conf.getInt(MRJobConfig.NUM_MAPS, 1));

    job.setJobName("Import job");
    job.setJarByClass(DBQueryInputJob.class);

    String strategy = conf.get(HIHOConf.INPUT_OUTPUT_STRATEGY);
    OutputStrategyEnum os = OutputStrategyEnum.value(strategy);
    if (os == null) {
        throw new IllegalArgumentException("Wrong value of output strategy. Please correct");
    }
    if (os != OutputStrategyEnum.AVRO) {
        switch (os) {

        case DUMP: {
            // job.setMapperClass(DBImportMapper.class);
            break;
        }
        /*
         * case AVRO: { job.setMapperClass(DBInputAvroMapper.class); //
         * need avro in cp // job.setJarByClass(Schema.class); // need
         * jackson which is needed by avro - ugly! //
         * job.setJarByClass(ObjectMapper.class);
         * job.setMapOutputKeyClass(NullWritable.class);
         * job.setMapOutputValueClass(AvroValue.class);
         * job.setOutputKeyClass(NullWritable.class);
         * job.setOutputValueClass(AvroValue.class);
         * job.setOutputFormatClass(AvroOutputFormat.class);
         * 
         * AvroOutputFormat.setOutputPath(job, new
         * Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH))); break; }
         */
        case DELIMITED: {
            job.setMapperClass(DBInputDelimMapper.class);
            job.setMapOutputKeyClass(Text.class);
            job.setMapOutputValueClass(Text.class);
            job.setOutputKeyClass(Text.class);
            job.setOutputValueClass(Text.class);
            job.setOutputFormatClass(NoKeyOnlyValueOutputFormat.class);

            NoKeyOnlyValueOutputFormat.setOutputPath(job, new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH)));
        }
        case JSON: {
            // job.setMapperClass(DBImportJsonMapper.class);
            // job.setJarByClass(ObjectMapper.class);
            break;
        }
        default: {
            job.setMapperClass(DBInputDelimMapper.class);
            job.setMapOutputKeyClass(Text.class);
            job.setMapOutputValueClass(Text.class);
            job.setOutputKeyClass(Text.class);
            job.setOutputValueClass(Text.class);
            job.setOutputFormatClass(NoKeyOnlyValueOutputFormat.class);

            NoKeyOnlyValueOutputFormat.setOutputPath(job, new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH)));
            break;
        }
        }

        String inputQuery = conf.get(DBConfiguration.INPUT_QUERY);
        String inputBoundingQuery = conf.get(DBConfiguration.INPUT_BOUNDING_QUERY);
        logger.debug("About to set the params");
        DBQueryInputFormat.setInput(job, inputQuery, inputBoundingQuery, params);
        logger.debug("Set the params");

        job.setNumReduceTasks(0);

        try {
            // job.setJarByClass(Class.forName(conf.get(
            // org.apache.hadoop.mapred.lib.db.DBConfiguration.DRIVER_CLASS_PROPERTY)));
            logger.debug("OUTPUT format class is " + job.getOutputFormatClass());

            /*
             * org.apache.hadoop.mapreduce.OutputFormat<?, ?> output =
             * ReflectionUtils.newInstance(job.getOutputFormatClass(),
             * job.getConfiguration()); output.checkOutputSpecs(job);
             */
            logger.debug("Class is " + ReflectionUtils
                    .newInstance(job.getOutputFormatClass(), job.getConfiguration()).getClass().getName());
            job.waitForCompletion(false);
            if (conf.get(HIHOConf.INPUT_OUTPUT_LOADTO) != null) {
                generateHiveScript(conf, job, jobCounter);
                generatePigScript(conf, job);
            }

        }
        /*
         * catch (HIHOException h) { h.printStackTrace(); }
         */
        catch (Exception e) {
            e.printStackTrace();
        } catch (HIHOException e) {
            e.printStackTrace();
        }
    }
    // avro to be handled differently, thanks to all the incompatibilities
    // in the apis.
    else {
        String inputQuery = conf.get(DBConfiguration.INPUT_QUERY);
        String inputBoundingQuery = conf.get(DBConfiguration.INPUT_BOUNDING_QUERY);
        logger.debug("About to set the params");
        // co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.setInput(job,
        // inputQuery, inputBoundingQuery, params);
        logger.debug("Set the params");

        JobConf jobConf = new JobConf(conf);

        try {
            GenericDBWritable queryWritable = getDBWritable(jobConf);
            Schema pair = DBMapper.getPairSchema(queryWritable.getColumns());

            AvroJob.setMapOutputSchema(jobConf, pair);
            GenericRecordAvroOutputFormat.setOutputPath(jobConf,
                    new Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH)));

            co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.setInput(jobConf, inputQuery,
                    inputBoundingQuery, params);
            jobConf.setInputFormat(co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.class);
            jobConf.setMapperClass(DBInputAvroMapper.class);
            jobConf.setMapOutputKeyClass(NullWritable.class);
            jobConf.setMapOutputValueClass(AvroValue.class);
            jobConf.setOutputKeyClass(NullWritable.class);
            jobConf.setOutputValueClass(Text.class);
            jobConf.setOutputFormat(GenericRecordAvroOutputFormat.class);
            jobConf.setJarByClass(DBQueryInputJob.class);
            jobConf.setStrings("io.serializations",
                    "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization,org.apache.avro.mapred.AvroSerialization");
            jobConf.setNumReduceTasks(0);
            /*
             * jobConf.setOutputFormat(org.apache.hadoop.mapred.
             * SequenceFileOutputFormat.class);
             * org.apache.hadoop.mapred.SequenceFileOutputFormat
             * .setOutputPath(jobConf, new
             * Path(getConf().get(HIHOConf.INPUT_OUTPUT_PATH)));
             */
            JobClient.runJob(jobConf);
        } catch (Throwable e) {
            e.printStackTrace();
        }

    }

}

From source file:com.alexholmes.hadooputils.combine.seqfile.mapred.CombineSequenceFileJob.java

License:Apache License

/**
 * The driver for the MapReduce job./*  w ww.  j  a  v  a  2  s .c o m*/
 *
 * @param conf           configuration
 * @param inputDirAsString  input directory in CSV-form
 * @param outputDirAsString output directory
 * @return true if the job completed successfully
 * @throws java.io.IOException         if something went wrong
 * @throws java.net.URISyntaxException if a URI wasn't correctly formed
 */
public boolean runJob(final Configuration conf, final String inputDirAsString, final String outputDirAsString)
        throws IOException, URISyntaxException, ClassNotFoundException, InterruptedException {

    JobConf job = new JobConf(conf);

    job.setJarByClass(CombineSequenceFileJob.class);
    job.setJobName("seqfilecombiner");

    job.setNumReduceTasks(0);

    job.setMapperClass(IdentityMapper.class);

    job.setInputFormat(CombineSequenceFileInputFormat.class);

    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(Text.class);

    FileInputFormat.setInputPaths(job, inputDirAsString);
    FileOutputFormat.setOutputPath(job, new Path(outputDirAsString));

    Date startTime = new Date();
    System.out.println("Job started: " + startTime);
    RunningJob jobResult = JobClient.runJob(job);

    Date endTime = new Date();
    System.out.println("Job ended: " + endTime);
    System.out.println("The job took "
            + TimeUnit.MILLISECONDS.toSeconds(endTime.getTime() - startTime.getTime()) + " seconds.");

    return jobResult.isSuccessful();
}