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:infinidb.hadoop.example.InfiniDoopDriver.java

License:Apache License

public int run(String[] args) throws Exception {
    Configuration conf = new Configuration();
    JobConf jobconf = new JobConf(conf, InfiniDoopDriver.class);
    DBConfiguration.configureDB(jobconf, "com.mysql.jdbc.Driver", "jdbc:mysql://srvswint4/tpch1", "root", "");
    String[] fields = { "n_nationkey", "n_name" };
    jobconf.setInputFormat(InfiniDBInputFormat.class);

    jobconf.setOutputKeyClass(LongWritable.class);
    jobconf.setOutputValueClass(Text.class);

    InfiniDBInputFormat.setInput(jobconf, InfiniDoopRecord.class, "nation", null, "n_nationkey", fields);

    InfiniDBConfiguration idbconf = new InfiniDBConfiguration(jobconf);
    idbconf.setOutputPath("output2");
    jobconf.setMapperClass(InfiniDoopInputMapper.class);
    jobconf.setNumMapTasks(4);//from www .j  a va2  s  . c om
    jobconf.setNumReduceTasks(1);
    jobconf.set("mapred.textoutputformat.separator", "|");
    JobClient client = new JobClient();

    client.setConf(jobconf);
    try {
        JobClient.runJob(jobconf);
    } catch (Exception e) {
        e.printStackTrace();
    }

    return 0;
}

From source file:io.aos.t4f.hadoop.mapred.WordCountMapReduceTest.java

License:Apache License

/**
 * The main driver for word count map/reduce program.
 * Invoke this method to submit the map/reduce job.
 * @throws IOException When there is communication problems with the 
 *                     job tracker./* ww w.  j  ava2 s  .  c  om*/
 */
public int run(String[] args) throws Exception {
    JobConf conf = new JobConf(getConf(), WordCountMapReduceTest.class);
    conf.setJobName("wordcount");

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

    conf.setMapperClass(WordCountMapper.class);
    conf.setCombinerClass(WordCountReducer.class);
    conf.setReducerClass(WordCountReducer.class);

    List<String> other_args = new ArrayList<String>();
    for (int i = 0; i < args.length; ++i) {
        try {
            if ("-m".equals(args[i])) {
                conf.setNumMapTasks(Integer.parseInt(args[++i]));
            } else if ("-r".equals(args[i])) {
                conf.setNumReduceTasks(Integer.parseInt(args[++i]));
            } else {
                other_args.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();
        }
    }
    // Make sure there are exactly 2 parameters left.
    if (other_args.size() != 2) {
        System.out.println("ERROR: Wrong number of parameters: " + other_args.size() + " instead of 2.");
        return printUsage();
    }
    FileInputFormat.setInputPaths(conf, other_args.get(0));
    FileOutputFormat.setOutputPath(conf, new Path(other_args.get(1)));

    JobClient.runJob(conf);
    return 0;
}

From source file:io.aos.t4f.hadoop.mapred.WordCountTest.java

License:Apache License

private JobConf createJobConf() {
    JobConf conf = mrCluster.createJobConf();
    conf.setJobName("wordcount test");

    conf.setMapperClass(WordCountMapper.class);
    conf.setReducerClass(WordCountReducer.class);

    conf.setInputFormat(TextInputFormat.class);
    conf.setMapOutputKeyClass(Text.class);
    conf.setMapOutputValueClass(IntWritable.class);
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(IntWritable.class);
    conf.setNumMapTasks(1);//  www.j a v a  2 s.co  m
    conf.setNumReduceTasks(1);
    FileInputFormat.setInputPaths(conf, input);
    FileOutputFormat.setOutputPath(conf, output);
    return conf;
}

From source file:io.bfscan.clueweb12.DumpWarcRecordsToPlainText.java

License:Apache License

/**
 * Runs this tool./*ww w.  j a  va  2  s .  co m*/
 */
@SuppressWarnings("static-access")
public int run(String[] args) throws Exception {
    Options options = new Options();

    options.addOption(
            OptionBuilder.withArgName("path").hasArg().withDescription("input path").create(INPUT_OPTION));
    options.addOption(
            OptionBuilder.withArgName("path").hasArg().withDescription("output path").create(OUTPUT_OPTION));

    CommandLine cmdline;
    CommandLineParser parser = new GnuParser();
    try {
        cmdline = parser.parse(options, args);
    } catch (ParseException exp) {
        HelpFormatter formatter = new HelpFormatter();
        formatter.printHelp(this.getClass().getName(), options);
        ToolRunner.printGenericCommandUsage(System.out);
        System.err.println("Error parsing command line: " + exp.getMessage());
        return -1;
    }

    if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(OUTPUT_OPTION)) {
        HelpFormatter formatter = new HelpFormatter();
        formatter.printHelp(this.getClass().getName(), options);
        ToolRunner.printGenericCommandUsage(System.out);
        return -1;
    }

    String input = cmdline.getOptionValue(INPUT_OPTION);
    String output = cmdline.getOptionValue(OUTPUT_OPTION);

    LOG.info("Tool name: " + DumpWarcRecordsToPlainText.class.getSimpleName());
    LOG.info(" - input: " + input);
    LOG.info(" - output: " + output);

    JobConf conf = new JobConf(getConf(), DumpWarcRecordsToPlainText.class);
    conf.setJobName(DumpWarcRecordsToPlainText.class.getSimpleName() + ":" + input);

    conf.setNumReduceTasks(0);

    FileInputFormat.addInputPaths(conf, input);
    FileOutputFormat.setOutputPath(conf, new Path(output));

    conf.setInputFormat(ClueWeb12InputFormat.class);
    conf.setOutputFormat(TextOutputFormat.class);
    conf.setMapperClass(MyMapper.class);

    RunningJob job = JobClient.runJob(conf);
    Counters counters = job.getCounters();
    int numDocs = (int) counters.findCounter(Records.PAGES).getCounter();

    LOG.info("Read " + numDocs + " docs.");

    return 0;
}

From source file:io.druid.indexer.updater.HadoopConverterJob.java

License:Apache License

public List<DataSegment> run() throws IOException {
    final JobConf jobConf = new JobConf();
    jobConf.setKeepFailedTaskFiles(false);
    for (Map.Entry<String, String> entry : converterConfig.getHadoopProperties().entrySet()) {
        jobConf.set(entry.getKey(), entry.getValue(), "converterConfig.getHadoopProperties()");
    }//from w  ww.java 2s.  c o  m
    final List<DataSegment> segments = converterConfig.getSegments();
    if (segments.isEmpty()) {
        throw new IAE("No segments found for datasource [%s]", converterConfig.getDataSource());
    }
    converterConfigIntoConfiguration(converterConfig, segments, jobConf);

    jobConf.setNumReduceTasks(0);// Map only. Number of map tasks determined by input format
    jobConf.setWorkingDirectory(new Path(converterConfig.getDistributedSuccessCache()));

    setJobName(jobConf, segments);

    if (converterConfig.getJobPriority() != null) {
        jobConf.setJobPriority(JobPriority.valueOf(converterConfig.getJobPriority()));
    }

    final Job job = Job.getInstance(jobConf);

    job.setInputFormatClass(ConfigInputFormat.class);
    job.setMapperClass(ConvertingMapper.class);
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(Text.class);
    job.setMapSpeculativeExecution(false);
    job.setOutputFormatClass(ConvertingOutputFormat.class);

    JobHelper.setupClasspath(JobHelper.distributedClassPath(jobConf.getWorkingDirectory()),
            JobHelper.distributedClassPath(getJobClassPathDir(job.getJobName(), jobConf.getWorkingDirectory())),
            job);

    Throwable throwable = null;
    try {
        job.submit();
        log.info("Job %s submitted, status available at %s", job.getJobName(), job.getTrackingURL());
        final boolean success = job.waitForCompletion(true);
        if (!success) {
            final TaskReport[] reports = job.getTaskReports(TaskType.MAP);
            if (reports != null) {
                for (final TaskReport report : reports) {
                    log.error("Error in task [%s] : %s", report.getTaskId(),
                            Arrays.toString(report.getDiagnostics()));
                }
            }
            return null;
        }
        try {
            loadedBytes = job.getCounters().findCounter(COUNTER_GROUP, COUNTER_LOADED).getValue();
            writtenBytes = job.getCounters().findCounter(COUNTER_GROUP, COUNTER_WRITTEN).getValue();
        } catch (IOException ex) {
            log.error(ex, "Could not fetch counters");
        }
        final JobID jobID = job.getJobID();

        final Path jobDir = getJobPath(jobID, job.getWorkingDirectory());
        final FileSystem fs = jobDir.getFileSystem(job.getConfiguration());
        final RemoteIterator<LocatedFileStatus> it = fs.listFiles(jobDir, true);
        final List<Path> goodPaths = new ArrayList<>();
        while (it.hasNext()) {
            final LocatedFileStatus locatedFileStatus = it.next();
            if (locatedFileStatus.isFile()) {
                final Path myPath = locatedFileStatus.getPath();
                if (ConvertingOutputFormat.DATA_SUCCESS_KEY.equals(myPath.getName())) {
                    goodPaths.add(new Path(myPath.getParent(), ConvertingOutputFormat.DATA_FILE_KEY));
                }
            }
        }
        if (goodPaths.isEmpty()) {
            log.warn("No good data found at [%s]", jobDir);
            return null;
        }
        final List<DataSegment> returnList = ImmutableList
                .copyOf(Lists.transform(goodPaths, new Function<Path, DataSegment>() {
                    @Nullable
                    @Override
                    public DataSegment apply(final Path input) {
                        try {
                            if (!fs.exists(input)) {
                                throw new ISE("Somehow [%s] was found but [%s] is missing at [%s]",
                                        ConvertingOutputFormat.DATA_SUCCESS_KEY,
                                        ConvertingOutputFormat.DATA_FILE_KEY, jobDir);
                            }
                        } catch (final IOException e) {
                            throw Throwables.propagate(e);
                        }
                        try (final InputStream stream = fs.open(input)) {
                            return HadoopDruidConverterConfig.jsonMapper.readValue(stream, DataSegment.class);
                        } catch (final IOException e) {
                            throw Throwables.propagate(e);
                        }
                    }
                }));
        if (returnList.size() == segments.size()) {
            return returnList;
        } else {
            throw new ISE(
                    "Tasks reported success but result length did not match! Expected %d found %d at path [%s]",
                    segments.size(), returnList.size(), jobDir);
        }
    } catch (InterruptedException | ClassNotFoundException e) {
        RuntimeException exception = Throwables.propagate(e);
        throwable = exception;
        throw exception;
    } catch (Throwable t) {
        throwable = t;
        throw t;
    } finally {
        try {
            cleanup(job);
        } catch (IOException e) {
            if (throwable != null) {
                throwable.addSuppressed(e);
            } else {
                log.error(e, "Could not clean up job [%s]", job.getJobID());
            }
        }
    }
}

From source file:io.fluo.stress.trie.Generate.java

License:Apache License

@Override
public int run(String[] args) throws Exception {

    if (args.length != 4) {
        log.error("Usage: " + this.getClass().getSimpleName()
                + " <numMappers> <numbersPerMapper> <max> <output dir>");
        System.exit(-1);// ww w . ja  v  a  2 s  .c o  m
    }

    int numMappers = Integer.parseInt(args[0]);
    int numPerMapper = Integer.parseInt(args[1]);
    long max = Long.parseLong(args[2]);
    Path out = new Path(args[3]);

    Preconditions.checkArgument(numMappers > 0, "numMappers <= 0");
    Preconditions.checkArgument(numPerMapper > 0, "numPerMapper <= 0");
    Preconditions.checkArgument(max > 0, "max <= 0");

    JobConf job = new JobConf(getConf());

    job.setJobName(this.getClass().getName());

    job.setJarByClass(Generate.class);

    job.setInt(TRIE_GEN_NUM_PER_MAPPER_PROP, numPerMapper);
    job.setInt(TRIE_GEN_NUM_MAPPERS_PROP, numMappers);
    job.setLong(TRIE_GEN_MAX_PROP, max);

    job.setInputFormat(RandomLongInputFormat.class);

    job.setNumReduceTasks(0);

    job.setOutputKeyClass(LongWritable.class);
    job.setOutputValueClass(NullWritable.class);

    job.setOutputFormat(SequenceFileOutputFormat.class);
    SequenceFileOutputFormat.setOutputPath(job, out);

    RunningJob runningJob = JobClient.runJob(job);
    runningJob.waitForCompletion();
    return runningJob.isSuccessful() ? 0 : -1;
}

From source file:io.hops.erasure_coding.MapReduceEncoder.java

License:Apache License

/**
 * create new job conf based on configuration passed.
 *
 * @param conf/*from   w w w.j  a v a  2 s.  c  o  m*/
 * @return
 */
private static JobConf createJobConf(Configuration conf) {
    JobConf jobconf = new JobConf(conf, MapReduceEncoder.class);
    jobName = NAME + " " + dateForm.format(new Date(BaseEncodingManager.now()));
    jobconf.setUser(BaseEncodingManager.JOBUSER);
    jobconf.setJobName(jobName);
    jobconf.setMapSpeculativeExecution(false);
    RaidUtils.parseAndSetOptions(jobconf, SCHEDULER_OPTION_LABEL);

    jobconf.setJarByClass(MapReduceEncoder.class);
    jobconf.setInputFormat(DistRaidInputFormat.class);
    jobconf.setOutputKeyClass(Text.class);
    jobconf.setOutputValueClass(Text.class);

    jobconf.setMapperClass(DistRaidMapper.class);
    jobconf.setNumReduceTasks(0);
    return jobconf;
}

From source file:it.crs4.pydoop.pipes.Submitter.java

License:Apache License

@Override
public int run(String[] args) throws Exception {
    CommandLineParser cli = new CommandLineParser();
    if (args.length == 0) {
        cli.printUsage();// w w  w . j a  v  a2 s . c o  m
        return 1;
    }
    cli.addOption("input", false, "input path to the maps", "path");
    cli.addOption("output", false, "output path from the reduces", "path");

    cli.addOption("jar", false, "job jar file", "path");
    cli.addOption("inputformat", false, "java classname of InputFormat", "class");
    //cli.addArgument("javareader", false, "is the RecordReader in Java");
    cli.addOption("map", false, "java classname of Mapper", "class");
    cli.addOption("partitioner", false, "java classname of Partitioner", "class");
    cli.addOption("reduce", false, "java classname of Reducer", "class");
    cli.addOption("writer", false, "java classname of OutputFormat", "class");
    cli.addOption("program", false, "URI to application executable", "class");
    cli.addOption("reduces", false, "number of reduces", "num");
    cli.addOption("jobconf", false,
            "\"n1=v1,n2=v2,..\" (Deprecated) Optional. Add or override a JobConf property.", "key=val");
    cli.addOption("lazyOutput", false, "Optional. Create output lazily", "boolean");
    Parser parser = cli.createParser();
    try {

        GenericOptionsParser genericParser = new GenericOptionsParser(getConf(), args);
        CommandLine results = parser.parse(cli.options, genericParser.getRemainingArgs());

        JobConf job = new JobConf(getConf());

        if (results.hasOption("input")) {
            FileInputFormat.setInputPaths(job, results.getOptionValue("input"));
        }
        if (results.hasOption("output")) {
            FileOutputFormat.setOutputPath(job, new Path(results.getOptionValue("output")));
        }
        if (results.hasOption("jar")) {
            job.setJar(results.getOptionValue("jar"));
        }
        if (results.hasOption("inputformat")) {
            setIsJavaRecordReader(job, true);
            job.setInputFormat(getClass(results, "inputformat", job, InputFormat.class));
        }
        if (results.hasOption("javareader")) {
            setIsJavaRecordReader(job, true);
        }
        if (results.hasOption("map")) {
            setIsJavaMapper(job, true);
            job.setMapperClass(getClass(results, "map", job, Mapper.class));
        }
        if (results.hasOption("partitioner")) {
            job.setPartitionerClass(getClass(results, "partitioner", job, Partitioner.class));
        }
        if (results.hasOption("reduce")) {
            setIsJavaReducer(job, true);
            job.setReducerClass(getClass(results, "reduce", job, Reducer.class));
        }
        if (results.hasOption("reduces")) {
            job.setNumReduceTasks(Integer.parseInt(results.getOptionValue("reduces")));
        }
        if (results.hasOption("writer")) {
            setIsJavaRecordWriter(job, true);
            job.setOutputFormat(getClass(results, "writer", job, OutputFormat.class));
        }

        if (results.hasOption("lazyOutput")) {
            if (Boolean.parseBoolean(results.getOptionValue("lazyOutput"))) {
                LazyOutputFormat.setOutputFormatClass(job, job.getOutputFormat().getClass());
            }
        }

        if (results.hasOption("program")) {
            setExecutable(job, results.getOptionValue("program"));
        }
        if (results.hasOption("jobconf")) {
            LOG.warn("-jobconf option is deprecated, please use -D instead.");
            String options = results.getOptionValue("jobconf");
            StringTokenizer tokenizer = new StringTokenizer(options, ",");
            while (tokenizer.hasMoreTokens()) {
                String keyVal = tokenizer.nextToken().trim();
                String[] keyValSplit = keyVal.split("=");
                job.set(keyValSplit[0], keyValSplit[1]);
            }
        }
        // if they gave us a jar file, include it into the class path
        String jarFile = job.getJar();
        if (jarFile != null) {
            final URL[] urls = new URL[] { FileSystem.getLocal(job).pathToFile(new Path(jarFile)).toURL() };
            //FindBugs complains that creating a URLClassLoader should be
            //in a doPrivileged() block. 
            ClassLoader loader = AccessController.doPrivileged(new PrivilegedAction<ClassLoader>() {
                public ClassLoader run() {
                    return new URLClassLoader(urls);
                }
            });
            job.setClassLoader(loader);
        }

        runJob(job);
        return 0;
    } catch (ParseException pe) {
        LOG.info("Error : " + pe);
        cli.printUsage();
        return 1;
    }

}

From source file:Iterator.SpeciesIterDriver2.java

@SuppressWarnings("deprecation")
public static void main(String[] args) {

    int iterationCount = 0;

    while (iterationCount <= 20) {

        System.out.println("Running Iteration - " + iterationCount);
        JobClient client = new JobClient();
        JobConf conf = new JobConf(SpeciesIterDriver2.class);
        conf.setJobName("Species Iter - " + iterationCount);

        // This property is set to generate 5 reducer tasks
        conf.setNumReduceTasks(5);
        conf.setOutputKeyClass(Text.class);
        conf.setOutputValueClass(Text.class);

        //output-iterator-0 contains the input data
        FileInputFormat.setInputPaths(conf, new Path("output-iterator-" + iterationCount));
        iterationCount++;//  w w w .java  2  s.  co m
        FileOutputFormat.setOutputPath(conf, new Path("output-iterator-" + iterationCount));

        conf.setMapperClass(SpeciesIterMapper2.class);
        conf.setReducerClass(SpeciesIterReducer2.class);
        conf.setCombinerClass(SpeciesIterReducer2.class);

        client.setConf(conf);
        try {
            JobClient.runJob(conf);
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

}

From source file:ivory.core.index.MergeGlobalStatsAcrossIndexSegments.java

License:Apache License

public int runTool() throws Exception {

    JobConf conf = new JobConf(getConf(), MergeGlobalStatsAcrossIndexSegments.class);
    FileSystem fs = FileSystem.get(conf);

    String collectionName = conf.get("Ivory.CollectionName");
    String indexPaths = conf.get("Ivory.IndexPaths");
    String dataOutputPath = conf.get("Ivory.DataOutputPath");
    int dfThreshold = conf.getInt("Ivory.DfThreshold", 0);

    // first, compute size of global term space
    Path tmpPaths = new Path("/tmp/index-paths.txt");

    FSDataOutputStream out = fs.create(tmpPaths, true);
    for (String s : indexPaths.split(",")) {
        out.write(new String(s + "\n").getBytes());
    }//from   w w  w . ja v  a 2  s . c  om
    out.close();

    LOG.info("Job: ComputeNumberOfTermsAcrossIndexSegments");
    conf.setJobName("ComputeNumberOfTermsAcrossIndexSegments:" + collectionName);

    FileInputFormat.addInputPath(conf, tmpPaths);

    conf.setNumMapTasks(1);
    conf.setNumReduceTasks(1);

    conf.set("mapred.child.java.opts", "-Xmx2048m");

    conf.setInputFormat(NLineInputFormat.class);
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(PairOfIntLong.class);
    conf.setOutputFormat(NullOutputFormat.class);

    conf.setMapperClass(MyMapper.class);
    conf.setReducerClass(IdentityReducer.class);

    long startTime = System.currentTimeMillis();
    RunningJob job = JobClient.runJob(conf);
    LOG.info("Job Finished in " + (System.currentTimeMillis() - startTime) / 1000.0 + " seconds");

    Counters counters = job.getCounters();

    long totalNumTerms = counters.findCounter("org.apache.hadoop.mapred.Task$Counter", 6, "REDUCE_INPUT_GROUPS")
            .getCounter();

    LOG.info("total number of terms in global dictionary = " + totalNumTerms);

    // now build the dictionary
    fs.delete(new Path(dataOutputPath), true);

    conf = new JobConf(getConf(), MergeGlobalStatsAcrossIndexSegments.class);

    LOG.info("Job: MergeGlobalStatsAcrossIndexSegments");
    conf.setJobName("MergeGlobalStatsAcrossIndexSegments:" + collectionName);

    FileInputFormat.addInputPath(conf, tmpPaths);

    conf.setNumMapTasks(1);
    conf.setNumReduceTasks(1);

    conf.set("mapred.child.java.opts", "-Xmx2048m");

    conf.setInputFormat(NLineInputFormat.class);
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(PairOfIntLong.class);
    conf.setOutputFormat(NullOutputFormat.class);

    conf.setMapperClass(MyMapper.class);
    conf.setReducerClass(MyReducer.class);

    conf.setLong("Ivory.IndexNumberOfTerms", (int) totalNumTerms);

    startTime = System.currentTimeMillis();
    job = JobClient.runJob(conf);
    LOG.info("Job Finished in " + (System.currentTimeMillis() - startTime) / 1000.0 + " seconds");

    // compute some # docs, collection length, avg doc length
    long collectionLength = 0;
    int docCount = 0;
    for (String index : indexPaths.split(",")) {
        LOG.info("reading stats for " + index);

        RetrievalEnvironment env = new RetrievalEnvironment(index, fs);

        long l = env.readCollectionLength();
        int n = env.readCollectionDocumentCount();

        LOG.info(" - CollectionLength: " + l);
        LOG.info(" - CollectionDocumentCount: " + n);

        collectionLength += l;
        docCount += n;
    }

    float avgdl = (float) collectionLength / docCount;

    LOG.info("all index segments: ");
    LOG.info(" - CollectionLength: " + collectionLength);
    LOG.info(" - CollectionDocumentCount: " + docCount);
    LOG.info(" - AverageDocumentLenght: " + avgdl);

    RetrievalEnvironment env = new RetrievalEnvironment(dataOutputPath, fs);

    env.writeCollectionAverageDocumentLength(avgdl);
    env.writeCollectionLength(collectionLength);
    env.writeCollectionDocumentCount(docCount);

    return 0;
}