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

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

Introduction

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

Prototype

public void setOutputFormat(Class<? extends OutputFormat> theClass) 

Source Link

Document

Set the OutputFormat implementation for the map-reduce job.

Usage

From source file:junto.algorithm.parallel.LP_ZGL_Hadoop.java

License:Apache License

public static void main(String[] args) throws Exception {
    Hashtable config = ConfigReader.read_config(args);

    String baseInputFilePat = Defaults.GetValueOrDie(config, "hdfs_input_pattern");
    String baseOutputFilePat = Defaults.GetValueOrDie(config, "hdfs_output_base");
    int numIterations = Integer.parseInt(Defaults.GetValueOrDie(config, "iters"));

    String currInputFilePat = baseInputFilePat;
    String currOutputFilePat = "";
    for (int iter = 1; iter <= numIterations; ++iter) {
        JobConf conf = new JobConf(LP_ZGL_Hadoop.class);
        conf.setJobName("lp_zgl_hadoop");

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

        conf.setMapperClass(LP_ZGL_Map.class);
        // conf.setCombinerClass(LP_ZGL_Reduce.class);
        conf.setReducerClass(LP_ZGL_Reduce.class);

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

        // hyperparameters
        conf.set("mu1", Defaults.GetValueOrDie(config, "mu1"));
        conf.set("mu2", Defaults.GetValueOrDie(config, "mu2"));
        conf.set("keepTopKLabels", Defaults.GetValueOrDefault((String) config.get("keep_top_k_labels"),
                Integer.toString(Integer.MAX_VALUE)));

        if (iter > 1) {
            // output from last iteration is the input for current iteration
            currInputFilePat = currOutputFilePat + "/*";
        }/*from ww w  .ja  v a  2s .c o  m*/
        FileInputFormat.setInputPaths(conf, new Path(currInputFilePat));

        currOutputFilePat = baseOutputFilePat + "_" + iter;
        FileOutputFormat.setOutputPath(conf, new Path(currOutputFilePat));

        JobClient.runJob(conf);
    }
}

From source file:junto.algorithm.parallel.MADHadoop.java

License:Apache License

public static void main(String[] args) throws Exception {
    Hashtable config = ConfigReader.read_config(args);

    String baseInputFilePat = Defaults.GetValueOrDie(config, "hdfs_input_pattern");
    String baseOutputFilePat = Defaults.GetValueOrDie(config, "hdfs_output_base");
    int numIterations = Integer.parseInt(Defaults.GetValueOrDie(config, "iters"));
    int numReducers = Defaults.GetValueOrDefault((String) config.get("num_reducers"), 10);

    String currInputFilePat = baseInputFilePat;
    String currOutputFilePat = "";
    for (int iter = 1; iter <= numIterations; ++iter) {
        JobConf conf = new JobConf(MADHadoop.class);
        conf.setJobName("mad_hadoop");

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

        conf.setMapperClass(MADHadoopMap.class);
        // conf.setCombinerClass(MADHadoopReduce.class);
        conf.setReducerClass(MADHadoopReduce.class);
        conf.setNumReduceTasks(numReducers);

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

        // hyperparameters
        conf.set("mu1", Defaults.GetValueOrDie(config, "mu1"));
        conf.set("mu2", Defaults.GetValueOrDie(config, "mu2"));
        conf.set("mu3", Defaults.GetValueOrDie(config, "mu3"));
        conf.set("keepTopKLabels", Defaults.GetValueOrDefault((String) config.get("keep_top_k_labels"),
                Integer.toString(Integer.MAX_VALUE)));

        if (iter > 1) {
            // output from last iteration is the input for current iteration
            currInputFilePat = currOutputFilePat + "/*";
        }//from  w  ww.  j  a  v a 2  s.  co  m
        FileInputFormat.setInputPaths(conf, new Path(currInputFilePat));

        currOutputFilePat = baseOutputFilePat + "_iter_" + iter;
        FileOutputFormat.setOutputPath(conf, new Path(currOutputFilePat));

        JobClient.runJob(conf);
    }
}

From source file:kafka.etl.impl.SimpleKafkaETLJob.java

License:Apache License

protected JobConf createJobConf() throws Exception {
    JobConf jobConf = KafkaETLJob.createJobConf("SimpleKafakETL", _topic, _props, getClass());

    jobConf.setMapperClass(SimpleKafkaETLMapper.class);
    KafkaETLInputFormat.setInputPaths(jobConf, new Path(_input));

    jobConf.setOutputKeyClass(LongWritable.class);
    jobConf.setOutputValueClass(Text.class);
    jobConf.setOutputFormat(TextOutputFormat.class);
    TextOutputFormat.setCompressOutput(jobConf, false);
    Path output = new Path(_output);
    FileSystem fs = output.getFileSystem(jobConf);
    if (fs.exists(output))
        fs.delete(output);/*from ww w  .ja  v  a  2s . co  m*/
    TextOutputFormat.setOutputPath(jobConf, output);

    jobConf.setNumReduceTasks(0);
    return jobConf;
}

From source file:mapreduce.DosAttack.java

License:Apache License

private void issue() throws IOException {
    LOG.info("Starting DOS on url[{}] with clients[{}]", wsURL, numMappers);
    DosMapper.init(wsURL);//from w ww. j a  v a 2  s.  c o  m
    JobConf job = new JobConf(DosAttack.class);
    job.setJarByClass(DosAttack.class);
    job.setJobName("DOS Attack");
    job.setNumReduceTasks(0);
    job.setInputFormat(NullInputFormat.class);
    job.setOutputFormat(NullOutputFormat.class);
    job.setMapperClass(DosMapper.class);
    job.setMapOutputKeyClass(NullWritable.class);
    job.setMapOutputValueClass(NullWritable.class);
    job.setOutputKeyClass(NullWritable.class);
    job.setOutputValueClass(NullWritable.class);
    job.setNumMapTasks(numMappers);
    job.setInt(NUM_MAPPERS_KEY, numMappers);
    job.setInt(NUM_REQUESTS_KEY, numRequests);
    job.set(TARGET_URL_KEY, wsURL);
    JobClient.runJob(job);
}

From source file:map_reduce.MapReduce_OptimizedBrandesAdditions_DO_JUNG.java

License:Open Source License

@SuppressWarnings("deprecation")
@Override/*from  w  w w. jav  a2  s  . c o m*/
public int run(String[] args) throws Exception {
    if (args.length < 1) {
        System.err.println("Usage:\n");
        System.exit(1);
    }

    //       Job job = new Job(super.getConf());

    //      READ IN ALL COMMAND LINE ARGUMENTS
    //      EXAMPLE: 
    // hadoop jar MapReduce_OptimizedBrandesAdditions_DO_JUNG.jar
    // -libjars collections-generic-4.01.jar,jung-graph-impl-2.0.1.jar,jung-api-2.0.1.jar
    // -Dmapred.job.map.memory.mb=4096
    // -Dmapred.job.reduce.memory.mb=4096
    // -Dmapred.child.java.opts=-Xmx3500m
    // -Dmapreduce.task.timeout=60000000
    // -Dmapreduce.job.queuename=QUEUENAME
    // input_iterbrandes_additions_nocomb_10k_1 output_iterbrandes_additions_nocomb_10k_1
    // 10 1 10000 55245 10k 10k_randedges 100 1 false times/ betweenness/

    int m = -1;

    // input path to use on hdfs
    Path inputPath = new Path(args[++m]);

    // output path to use on hdfs
    Path outputPath = new Path(args[++m]);

    // number of Mappers to split the sources: e.g., 1, 10, 100 etc.
    // rule of thumb: the larger the graph (i.e., number of roots to test), the larger should be this number.
    int numOfMaps = Integer.parseInt(args[++m]);

    // number of Reducers to collect the output
    int numOfReduce = Integer.parseInt(args[++m]);

    // Number of vertices in graph
    int N = Integer.parseInt(args[++m]);

    // Number of edges in graph
    int M = Integer.parseInt(args[++m]);

    // Graph file (edge list, tab delimited) (full path)
    String graph = args[++m];

    // File with edges to be added (tab delimited) (full path)
    // Note: this version handles only edges between existing vertices in the graph.
    String random_edges = args[++m];

    // Number of random edges added
    int re = Integer.parseInt(args[++m]);

    // Experiment iteration (in case of multiple experiments)
    int iter = Integer.parseInt(args[++m]);

    // Use combiner or not (true/false)
    Boolean comb = Boolean.valueOf(args[++m]);

    // Output path for file with stats
    String statsoutputpath = args[++m];

    // Output path for file with final betweenness values
    String betoutputpath = args[++m];

    //      BEGIN INITIALIZATION

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

    String setup = "_additions_edges" + re + "_maps" + numOfMaps + "_comb" + comb;
    conf.setJobName("OptimizedBrandesAdditionsDOJung_" + graph + setup + "_" + iter);
    conf.set("HDFS_GRAPH", graph + setup);
    conf.set("HDFS_Random_Edges", random_edges + setup);
    conf.set("output", outputPath.getName());
    conf.set("setup", setup);

    //      CREATE INPUT FILES FOR MAPPERS

    int numOfTasksperMap = (int) Math.ceil(N / numOfMaps);
    //generate an input file for each map task
    for (int i = 0; i < numOfMaps - 1; i++) {
        Path file = new Path(inputPath, "part-r-" + i);
        IntWritable start = new IntWritable(i * numOfTasksperMap);
        IntWritable end = new IntWritable((i * numOfTasksperMap) + numOfTasksperMap - 1);

        SequenceFile.Writer writer = SequenceFile.createWriter(fs, conf, file, IntWritable.class,
                IntWritable.class, CompressionType.NONE);
        try {
            writer.append(start, end);
        } finally {
            writer.close();
        }
        System.out.println("Wrote input for Map #" + i + ": " + start + " - " + end);
    }

    // last mapper takes what is left
    Path file = new Path(inputPath, "part-r-" + (numOfMaps - 1));
    IntWritable start = new IntWritable((numOfMaps - 1) * numOfTasksperMap);
    IntWritable end = new IntWritable(N - 1);
    SequenceFile.Writer writer = SequenceFile.createWriter(fs, conf, file, IntWritable.class, IntWritable.class,
            CompressionType.NONE);
    try {
        writer.append(start, end);
    } finally {
        writer.close();
    }
    System.out.println("Wrote input for Map #" + (numOfMaps - 1) + ": " + start + " - " + end);

    //      COPY FILES TO MAPPERS
    System.out.println("Copying graph to cache");
    String LOCAL_GRAPH = graph;
    Path hdfsPath = new Path(graph + setup);

    // upload the file to hdfs. Overwrite any existing copy.
    fs.copyFromLocalFile(false, true, new Path(LOCAL_GRAPH), hdfsPath);
    DistributedCache.addCacheFile(hdfsPath.toUri(), conf);

    System.out.println("Copying random edges to cache");
    String LOCAL_Random_Edges = random_edges;
    hdfsPath = new Path(random_edges + setup);

    // upload the file to hdfs. Overwrite any existing copy.
    fs.copyFromLocalFile(false, true, new Path(LOCAL_Random_Edges), hdfsPath);
    DistributedCache.addCacheFile(hdfsPath.toUri(), conf);

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

    conf.setMapperClass(IterBrandesMapper.class);
    conf.setNumMapTasks(numOfMaps);

    if (comb)
        conf.setCombinerClass(IterBrandesReducer.class);

    conf.setReducerClass(IterBrandesReducer.class);
    conf.setNumReduceTasks(numOfReduce);

    // turn off speculative execution, because DFS doesn't handle multiple writers to the same file.
    conf.setSpeculativeExecution(false);

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);

    FileInputFormat.setInputPaths(conf, inputPath);
    FileOutputFormat.setOutputPath(conf, outputPath);

    // conf.set("mapred.job.name", "APS-" + outputPath.getName());
    conf.setNumTasksToExecutePerJvm(-1); // JVM reuse

    System.out.println("Starting the execution...! Pray!! \n");
    long time1 = System.nanoTime();
    RunningJob rj = JobClient.runJob(conf);
    long time2 = System.nanoTime();

    //      READ OUTPUT FILES

    System.out.println("\nFinished and now reading/writing Betweenness Output...\n");

    // Assuming 1 reducer.
    Path inFile = new Path(outputPath, "part-00000");
    IntWritable id = new IntWritable();
    DoubleWritable betweenness = new DoubleWritable();
    SequenceFile.Reader reader = new SequenceFile.Reader(fs, inFile, conf);

    FileWriter fw = new FileWriter(new File(betoutputpath + graph + setup + "_betweenness_" + iter));
    try {
        int i = 0;
        for (; i < (N + M + re); i++) {
            reader.next(id, betweenness);
            fw.write(id + "\t" + betweenness + "\n");
            fw.flush();
        }
    } finally {
        reader.close();
        fw.close();
    }

    System.out.println("\nWriting times Output...\n");

    fw = new FileWriter(new File(statsoutputpath + graph + setup + "_times_" + iter));

    fw.write("Total-time:\t" + (time2 - time1) + "\n");
    fw.write("total-map\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("SLOTS_MILLIS_MAPS") + "\n");
    fw.write("total-reduce\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("SLOTS_MILLIS_REDUCES") + "\n");
    fw.write("total-cpu-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("CPU_MILLISECONDS") + "\n");
    fw.write("total-gc-mr\t"
            + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter").getCounter("GC_TIME_MILLIS")
            + "\n");
    fw.write("total-phy-mem-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("PHYSICAL_MEMORY_BYTES") + "\n");
    fw.write("total-vir-mem-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("VIRTUAL_MEMORY_BYTES") + "\n");
    fw.write("brandes\t" + rj.getCounters().getGroup("TimeForBrandes").getCounter("exectime_initial_brandes")
            + "\n");
    fw.write("reduce\t" + rj.getCounters().getGroup("TimeForReduce").getCounter("reduceafteralledges") + "\n");
    fw.flush();

    try {
        Iterator<Counters.Counter> counters = rj.getCounters().getGroup("TimeForRandomEdges").iterator();
        while (counters.hasNext()) {
            Counter cc = counters.next();
            fw.write(cc.getName() + "\t" + cc.getCounter() + "\n");
            fw.flush();
        }
    } finally {
        fw.close();
    }

    return 0;
}

From source file:map_reduce.MapReduce_OptimizedBrandesDeletions_DO_JUNG.java

License:Open Source License

@SuppressWarnings("deprecation")
@Override//from   w w  w . ja  v  a 2s  .c  o m
public int run(String[] args) throws Exception {
    if (args.length < 1) {
        System.err.println("Usage:\n");
        System.exit(1);
    }

    //       Job job = new Job(super.getConf());

    //      READ IN ALL COMMAND LINE ARGUMENTS
    //      EXAMPLE: 
    // hadoop jar MapReduce_OptimizedBrandesDeletions_DO_JUNG.jar
    // -libjars collections-generic-4.01.jar,jung-graph-impl-2.0.1.jar,jung-api-2.0.1.jar
    // -Dmapred.job.map.memory.mb=4096
    // -Dmapred.job.reduce.memory.mb=4096
    // -Dmapred.child.java.opts=-Xmx3500m
    // -Dmapreduce.task.timeout=60000000
    // -Dmapreduce.job.queuename=QUEUENAME
    // input_iterbrandes_deletions_nocomb_10k_1 output_iterbrandes_deletions_nocomb_10k_1
    // 10 1 10000 55245 10k 10k_randedges 100 1 false times/ betweenness/

    int m = -1;

    // input path to use on hdfs
    Path inputPath = new Path(args[++m]);

    // output path to use on hdfs
    Path outputPath = new Path(args[++m]);

    // number of Mappers to split the sources: e.g., 1, 10, 100 etc.
    // rule of thumb: the larger the graph (i.e., number of roots to test), the larger should be this number.
    int numOfMaps = Integer.parseInt(args[++m]);

    // number of Reducers to collect the output
    int numOfReduce = Integer.parseInt(args[++m]);

    // Number of vertices in graph
    int N = Integer.parseInt(args[++m]);

    // Number of edges in graph
    int M = Integer.parseInt(args[++m]);

    // Graph file (edge list, tab delimited) (full path)
    String graph = args[++m];

    // File with edges to be added (tab delimited) (full path)
    // Note: this version handles only edges between existing vertices in the graph.
    String random_edges = args[++m];

    // Number of random edges added
    int re = Integer.parseInt(args[++m]);

    // Experiment iteration (in case of multiple experiments)
    int iter = Integer.parseInt(args[++m]);

    // Use combiner or not (true/false)
    Boolean comb = Boolean.valueOf(args[++m]);

    // Output path for file with stats
    String statsoutputpath = args[++m];

    // Output path for file with final betweenness values
    String betoutputpath = args[++m];

    //      BEGIN INITIALIZATION

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

    String setup = "_deletions_edges" + re + "_maps" + numOfMaps + "_comb" + comb;
    conf.setJobName("OptimizedBrandesDeletionsDOJung_" + graph + setup + "_" + iter);
    conf.set("HDFS_GRAPH", graph + setup);
    conf.set("HDFS_Random_Edges", random_edges + setup);
    conf.set("output", outputPath.getName());
    conf.set("setup", setup);

    //      CREATE INPUT FILES FOR MAPPERS

    int numOfTasksperMap = (int) Math.ceil(N / numOfMaps);
    //generate an input file for each map task
    for (int i = 0; i < numOfMaps - 1; i++) {
        Path file = new Path(inputPath, "part-r-" + i);
        IntWritable start = new IntWritable(i * numOfTasksperMap);
        IntWritable end = new IntWritable((i * numOfTasksperMap) + numOfTasksperMap - 1);

        SequenceFile.Writer writer = SequenceFile.createWriter(fs, conf, file, IntWritable.class,
                IntWritable.class, CompressionType.NONE);
        try {
            writer.append(start, end);
        } finally {
            writer.close();
        }
        System.out.println("Wrote input for Map #" + i + ": " + start + " - " + end);
    }

    // last mapper takes what is left
    Path file = new Path(inputPath, "part-r-" + (numOfMaps - 1));
    IntWritable start = new IntWritable((numOfMaps - 1) * numOfTasksperMap);
    IntWritable end = new IntWritable(N - 1);
    SequenceFile.Writer writer = SequenceFile.createWriter(fs, conf, file, IntWritable.class, IntWritable.class,
            CompressionType.NONE);
    try {
        writer.append(start, end);
    } finally {
        writer.close();
    }
    System.out.println("Wrote input for Map #" + (numOfMaps - 1) + ": " + start + " - " + end);

    //      COPY FILES TO MAPPERS
    System.out.println("Copying graph to cache");
    String LOCAL_GRAPH = graph;
    Path hdfsPath = new Path(graph + setup);

    // upload the file to hdfs. Overwrite any existing copy.
    fs.copyFromLocalFile(false, true, new Path(LOCAL_GRAPH), hdfsPath);
    DistributedCache.addCacheFile(hdfsPath.toUri(), conf);

    System.out.println("Copying random edges to cache");
    String LOCAL_Random_Edges = random_edges;
    hdfsPath = new Path(random_edges + setup);

    // upload the file to hdfs. Overwrite any existing copy.
    fs.copyFromLocalFile(false, true, new Path(LOCAL_Random_Edges), hdfsPath);
    DistributedCache.addCacheFile(hdfsPath.toUri(), conf);

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

    conf.setMapperClass(IterBrandesMapper.class);
    conf.setNumMapTasks(numOfMaps);

    if (comb)
        conf.setCombinerClass(IterBrandesReducer.class);

    conf.setReducerClass(IterBrandesReducer.class);
    conf.setNumReduceTasks(numOfReduce);

    // turn off speculative execution, because DFS doesn't handle multiple writers to the same file.
    conf.setSpeculativeExecution(false);

    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);

    FileInputFormat.setInputPaths(conf, inputPath);
    FileOutputFormat.setOutputPath(conf, outputPath);

    // conf.set("mapred.job.name", "APS-" + outputPath.getName());
    conf.setNumTasksToExecutePerJvm(-1); // JVM reuse

    System.out.println("Starting the execution...! Pray!! \n");
    long time1 = System.nanoTime();
    RunningJob rj = JobClient.runJob(conf);
    long time2 = System.nanoTime();

    //      READ OUTPUT FILES

    System.out.println("\nFinished and now reading/writing Betweenness Output...\n");

    // Assuming 1 reducer.
    Path inFile = new Path(outputPath, "part-00000");
    IntWritable id = new IntWritable();
    DoubleWritable betweenness = new DoubleWritable();
    SequenceFile.Reader reader = new SequenceFile.Reader(fs, inFile, conf);

    FileWriter fw = new FileWriter(new File(betoutputpath + graph + setup + "_betweenness_" + iter));
    try {
        int i = 0;
        for (; i < (N + (M - re)); i++) {
            reader.next(id, betweenness);
            fw.write(id + "\t" + betweenness + "\n");
            fw.flush();
        }
    } finally {
        reader.close();
        fw.close();
    }

    System.out.println("\nWriting times Output...\n");

    fw = new FileWriter(new File(statsoutputpath + graph + setup + "_times_" + iter));

    fw.write("Total-time:\t" + (time2 - time1) + "\n");
    fw.write("total-map\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("SLOTS_MILLIS_MAPS") + "\n");
    fw.write("total-reduce\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("SLOTS_MILLIS_REDUCES") + "\n");
    fw.write("total-cpu-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("CPU_MILLISECONDS") + "\n");
    fw.write("total-gc-mr\t"
            + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter").getCounter("GC_TIME_MILLIS")
            + "\n");
    fw.write("total-phy-mem-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("PHYSICAL_MEMORY_BYTES") + "\n");
    fw.write("total-vir-mem-mr\t" + rj.getCounters().getGroup("org.apache.hadoop.mapreduce.TaskCounter")
            .getCounter("VIRTUAL_MEMORY_BYTES") + "\n");
    fw.write("brandes\t" + rj.getCounters().getGroup("TimeForBrandes").getCounter("exectime_initial_brandes")
            + "\n");
    fw.write("reduce\t" + rj.getCounters().getGroup("TimeForReduce").getCounter("reduceafteralledges") + "\n");
    fw.flush();

    try {
        Iterator<Counters.Counter> counters = rj.getCounters().getGroup("TimeForRandomEdges").iterator();
        while (counters.hasNext()) {
            Counter cc = counters.next();
            fw.write(cc.getName() + "\t" + cc.getCounter() + "\n");
            fw.flush();
        }
    } finally {
        fw.close();
    }

    return 0;
}

From source file:mr.WordCount.java

License:Open Source License

public static void main(String[] args) throws Exception {

    Properties properties = new Properties();
    AppProps.addApplicationTag(properties, "tutorials");
    AppProps.addApplicationTag(properties, "cluster:development");
    AppProps.setApplicationName(properties, "cascading-mapreduce-flow");

    JobConf conf = new JobConf(WordCount.class);
    conf.setJobName("casading-mapreduce-flow");

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

    conf.setMapperClass(Map.class);
    conf.setCombinerClass(Reduce.class);
    conf.setReducerClass(Reduce.class);

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

    FileInputFormat.setInputPaths(conf, new Path(args[0]));
    FileOutputFormat.setOutputPath(conf, new Path(args[1]));

    MapReduceFlow flow = new MapReduceFlow("wordcount", conf, true);

    // JobClient.runJob(conf);
    flow.complete();// w w w  .jav  a  2 s . c  o  m

}

From source file:name.abhijitsarkar.hadoop.citation.CitationCombiner.java

License:Open Source License

@Override
public int run(String[] args) throws Exception {
    JobConf conf = new JobConf(getConf(), getClass());
    conf.setJobName("citation-combiner");

    /* This is to set the separator byte for KeyValueTextInputFormat */
    conf.set("key.value.separator.in.input.line", ",");

    conf.setMapperClass(CitationMapper.class);
    conf.setReducerClass(CitationReducer.class);

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

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

    FileInputFormat.setInputPaths(conf, new Path(args[0]));
    FileOutputFormat.setOutputPath(conf, new Path(args[1]));

    JobClient.runJob(conf);/*from  w w w . jav a  2s .  c  o  m*/

    return 0;
}

From source file:name.abhijitsarkar.hadoop.citation.CitationCombinerWithChaining.java

License:Open Source License

@Override
public int run(String[] args) throws Exception {
    JobConf conf = new JobConf(getConf(), getClass());
    conf.setJobName("citation-combiner-with-chaining");

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

    FileInputFormat.setInputPaths(conf, new Path(args[0]));
    FileOutputFormat.setOutputPath(conf, new Path(args[1]));

    JobConf map1Conf = new JobConf(false);

    ChainMapper.addMapper(conf, CitationInputSplitMapper.class, LongWritable.class, Text.class, Text.class,
            Text.class, true, map1Conf);

    JobConf map2Conf = new JobConf(false);

    ChainMapper.addMapper(conf, CitationHeaderStripMapper.class, Text.class, Text.class, Text.class, Text.class,
            true, map2Conf);/*from  w w w.  j  ava 2  s  .  c  o  m*/

    JobConf red1Conf = new JobConf(false);

    ChainReducer.setReducer(conf, CitationReducer.class, Text.class, Text.class, Text.class, Text.class, true,
            red1Conf);

    JobClient.runJob(conf);

    return 0;
}

From source file:NCDSearch.DistributedNCDSearch.java

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

    String inputpath = args[1];/*from  w  ww .ja v  a2  s  . c o  m*/
    String outputpath = args[2];

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

    //Add the target file to a cache so all nodes can have a copy.
    DistributedCache.addCacheFile(new URI(args[0]), conf);

    FileOutputFormat.setOutputPath(conf, new Path(outputpath));
    FileInputFormat.setInputPaths(conf, new Path(inputpath));

    conf.setJobName("NCDSearch");
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(FloatWritable.class);
    conf.setOutputFormat(TextOutputFormat.class);

    conf.setInputFormat(ChunkyFileInputFormat.class);
    conf.setMapperClass(Map.class);
    conf.setReducerClass(Reduce.class);

    JobClient.runJob(conf);

    return 0;
}