Package org.apache.hadoop.mapred

Examples of org.apache.hadoop.mapred.JobConf

310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
    /**
     * Run the job
     */
    public void build() {
        try {
            JobConf conf = new JobConf(config);
            conf.setInt("io.file.buffer.size", DEFAULT_BUFFER_SIZE);
            conf.set("cluster.xml", new ClusterMapper().writeCluster(cluster));
            conf.set("stores.xml",
                     new StoreDefinitionsMapper().writeStoreList(Collections.singletonList(storeDef)));
            conf.setBoolean("save.keys", saveKeys);
            conf.setBoolean("reducer.per.bucket", reducerPerBucket);
            if(!isAvro) {
                conf.setPartitionerClass(HadoopStoreBuilderPartitioner.class);
                conf.setMapperClass(mapperClass);
                conf.setMapOutputKeyClass(BytesWritable.class);
                conf.setMapOutputValueClass(BytesWritable.class);
                if(reducerPerBucket) {
                    conf.setReducerClass(HadoopStoreBuilderReducerPerBucket.class);
                } else {
                    conf.setReducerClass(HadoopStoreBuilderReducer.class);
                }
            }
            conf.setInputFormat(inputFormatClass);
            conf.setOutputFormat(SequenceFileOutputFormat.class);
            conf.setOutputKeyClass(BytesWritable.class);
            conf.setOutputValueClass(BytesWritable.class);
            conf.setJarByClass(getClass());
            conf.setReduceSpeculativeExecution(false);
            FileInputFormat.setInputPaths(conf, inputPath);
            conf.set("final.output.dir", outputDir.toString());
            conf.set("checksum.type", CheckSum.toString(checkSumType));
            FileOutputFormat.setOutputPath(conf, tempDir);

            FileSystem outputFs = outputDir.getFileSystem(conf);
            if(outputFs.exists(outputDir)) {
                throw new IOException("Final output directory already exists.");
            }

            // delete output dir if it already exists
            FileSystem tempFs = tempDir.getFileSystem(conf);
            tempFs.delete(tempDir, true);

            long size = sizeOfPath(tempFs, inputPath);
            logger.info("Data size = " + size + ", replication factor = "
                        + storeDef.getReplicationFactor() + ", numNodes = "
                        + cluster.getNumberOfNodes() + ", chunk size = " + chunkSizeBytes);

            // Derive "rough" number of chunks and reducers
            int numReducers;
            if(saveKeys) {

                if(this.numChunks == -1) {
                    this.numChunks = Math.max((int) (storeDef.getReplicationFactor() * size
                                                     / cluster.getNumberOfPartitions()
                                                     / storeDef.getReplicationFactor() / chunkSizeBytes),
                                              1);
                } else {
                    logger.info("Overriding chunk size byte and taking num chunks ("
                                + this.numChunks + ") directly");
                }

                if(reducerPerBucket) {
                    numReducers = cluster.getNumberOfPartitions() * storeDef.getReplicationFactor();
                } else {
                    numReducers = cluster.getNumberOfPartitions() * storeDef.getReplicationFactor()
                                  * numChunks;
                }
            } else {

                if(this.numChunks == -1) {
                    this.numChunks = Math.max((int) (storeDef.getReplicationFactor() * size
                                                     / cluster.getNumberOfPartitions() / chunkSizeBytes),
                                              1);
                } else {
                    logger.info("Overriding chunk size byte and taking num chunks ("
                                + this.numChunks + ") directly");
                }

                if(reducerPerBucket) {
                    numReducers = cluster.getNumberOfPartitions();
                } else {
                    numReducers = cluster.getNumberOfPartitions() * numChunks;
                }
            }
            conf.setInt("num.chunks", numChunks);
            conf.setNumReduceTasks(numReducers);

            if(isAvro) {
                conf.setPartitionerClass(AvroStoreBuilderPartitioner.class);
                // conf.setMapperClass(mapperClass);
                conf.setMapOutputKeyClass(ByteBuffer.class);
                conf.setMapOutputValueClass(ByteBuffer.class);

                conf.setInputFormat(inputFormatClass);

                conf.setOutputFormat((Class<? extends OutputFormat>) AvroOutputFormat.class);
                conf.setOutputKeyClass(ByteBuffer.class);
                conf.setOutputValueClass(ByteBuffer.class);

                // AvroJob confs for the avro mapper
                AvroJob.setInputSchema(conf, Schema.parse(config.get("avro.rec.schema")));

                AvroJob.setOutputSchema(conf,
                                        Pair.getPairSchema(Schema.create(Schema.Type.BYTES),
                                                           Schema.create(Schema.Type.BYTES)));

                AvroJob.setMapperClass(conf, mapperClass);

                if(reducerPerBucket) {
                    conf.setReducerClass(AvroStoreBuilderReducerPerBucket.class);
                } else {
                    conf.setReducerClass(AvroStoreBuilderReducer.class);
                }

            }

            logger.info("Number of chunks: " + numChunks + ", number of reducers: " + numReducers
View Full Code Here
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
  @Override
  public RecordReader<NullWritable, NullWritable> getRecordReader(InputSplit arg0, JobConf arg1,
        Reporter arg2) throws IOException
  {
    final String filename = ((FileSplit)arg0).getPath().toString();
    final JobConf job = arg1;
   
    return new RecordReader<NullWritable, NullWritable>()
    {
      private boolean unread = true;

      @Override
      public void close() throws IOException
      {}

      @Override
      public NullWritable createKey()
      {
        return NullWritable.get();
      }

      @Override
      public NullWritable createValue()
      {
        return NullWritable.get();
      }

      @Override
      public long getPos() throws IOException
      {
        return 0;
      }

      @Override
      public float getProgress() throws IOException
      {
        return unread ? 0 : 1;
      }

      @Override
      /* spawn a cpimport process for each input file */
      public boolean next(NullWritable arg0, NullWritable arg1) throws IOException
      {
        InfiniDBConfiguration dbConf = new InfiniDBConfiguration(job);
        String schemaName = dbConf.getOutputSchemaName();
        String tableName = (filename.substring(filename.lastIndexOf('/')+1, filename.length()));
        tableName = tableName.substring(0, tableName.lastIndexOf('.'));
        String output = job.get("mapred.output.dir");
        if (unread)
        {
          try
          {
            StringBuilder loadCmdStr = new StringBuilder();
            loadCmdStr.append(dbConf.getInfiniDBHome());
            loadCmdStr.append("/bin/");
            loadCmdStr.append("infinidoop_load.sh ");
            loadCmdStr.append(filename);
            loadCmdStr.append(" ");
            loadCmdStr.append(schemaName);
            loadCmdStr.append(" ");
            loadCmdStr.append(tableName);

            Process lChldProc = Runtime.getRuntime().exec(loadCmdStr.toString());
            
            // Wait for the child to exit
            lChldProc.waitFor();
            BufferedReader lChldProcOutStream = new BufferedReader(new InputStreamReader(lChldProc.getInputStream()));
            BufferedReader stdError = new BufferedReader(new InputStreamReader(lChldProc.getErrorStream()));
           
            String lChldProcOutPutStr = null;
            StringBuffer outpath = new StringBuffer();
            outpath.append(job.getWorkingDirectory());
            outpath.append("/");
            outpath.append(output);
            outpath.append("/");
            outpath.append(tableName);
            outpath.append(".log");
View Full Code Here
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
public class InfiniDBOutputDriver extends Configured implements Tool
{
  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" };
    String [] outFields = {"id", "name"};
    jobconf.setInputFormat(IDBFileInputFormat.class);
    jobconf.setOutputFormat(InfiniDBOutputFormat.class);
    jobconf.setOutputKeyClass(NullWritable.class);
    jobconf.setOutputValueClass(Text.class);
    InfiniDBOutputFormat.setOutput(jobconf, "db", outFields);
    InfiniDBConfiguration idbconf = new InfiniDBConfiguration(jobconf);
    idbconf.setInputPath("input");
    idbconf.setOutputPath("output");
    idbconf.setInfiniDBHome("/usr/local/Calpont");

    jobconf.setMapperClass(InfiniDoopMapper.class);
    jobconf.setNumMapTasks(1);
    jobconf.setNumReduceTasks(2);
    JobClient client = new JobClient();
    client.setConf(jobconf);
    try {
      JobClient.runJob(jobconf);
    } catch (Exception e) {
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
public class InfiniDoopDriver extends Configured implements Tool
{
  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);
    jobconf.setNumReduceTasks(1);
    jobconf.set("mapred.textoutputformat.separator", "|");
    JobClient client = new JobClient();

    client.setConf(jobconf);
    try {
      JobClient.runJob(jobconf);
View Full Code Here
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
public class InfiniDBOutputDriver extends Configured implements Tool
{
  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" };
    String [] outFields = {"id", "name"};
    jobconf.setInputFormat(IDBFileInputFormat.class);
    jobconf.setOutputFormat(InfiniDBOutputFormat.class);
    //Class.forName(jobconf.get(DBConfiguration.DRIVER_CLASS_PROPERTY));
    //Connection connection = DriverManager.getConnection("jdbc:mysql://srvswint4/tpch1?user=root");
   
        //Job job = new Job(jobconf, "infinidb_driver");
        //jobconf.setOutputKeyClass(InfiniDoopRecord.class);
    jobconf.setOutputKeyClass(NullWritable.class);
        jobconf.setOutputValueClass(NullWritable.class);
        //jobconf.setMapOutputKeyClass(InfiniDoopRecord.class);
        //jobconf.setMapOutputValueClass(LongWritable.class);
       
        //InfiniDBInputFormat.setInput(jobconf, InfiniDoopRecord.class, "nation",
     //      null,  "n_nationkey", fields);
        InfiniDBOutputFormat.setOutput(jobconf, "t1", outFields);
        //InfiniDBInputFormat.setInput(jobconf, InfiniDBRecord.class, "select * from nation", "select count(*) from nation");
    Job job = new Job(jobconf);
    jobconf.set("mapred.input.dir", "output1");
    jobconf.set("mapred.output.dir", "output");
   
        jobconf.setMapperClass(InfiniDoopMapper.class);
        jobconf.setNumMapTasks(1);
        jobconf.setNumReduceTasks(2);
        JobClient client = new JobClient();
        //System.out.println(jobconf.getUser());
        //System.out.println(FileOutputFormat.getOutputPath(jobconf));
        jobconf.setInt("mapred.min.split.size", 2147483647);

        //int i = (job.waitForCompletion(true)? 1: 0);
        //return i;
        client.setConf(jobconf);
    try {
View Full Code Here
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
public class InfiniDoopDriver extends Configured implements Tool
{
  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" };
    String [] outFields = {"id", "name"};
    jobconf.setInputFormat(InfiniDBInputFormat.class);
    //Class.forName(jobconf.get(DBConfiguration.DRIVER_CLASS_PROPERTY));
    //Connection connection = DriverManager.getConnection("jdbc:mysql://srvswint4/tpch1?user=root");
 
        //Job job = new Job(jobconf, "infinidb_driver");
        jobconf.setOutputKeyClass(LongWritable.class);
        jobconf.setOutputValueClass(Text.class);
        //jobconf.setMapOutputKeyClass(LongWritable.class);
        //jobconf.setMapOutputValueClass(LongWritable.class);
       
        InfiniDBInputFormat.setInput(jobconf, InfiniDoopRecord.class, "nation",
           null,  "n_nationkey", fields);
       
        //InfiniDBInputFormat.setInput(jobconf, InfiniDBRecord.class, "select * from nation", "select count(*) from nation");
    Job job = new Job(jobconf);
    jobconf.set("mapred.output.dir", "output1");
        jobconf.setMapperClass(InfiniDoopInputMapper.class);
        jobconf.setNumMapTasks(4);
        jobconf.setNumReduceTasks(8);
        JobClient client = new JobClient();
        //System.out.println(jobconf.getUser());
        //System.out.println(FileOutputFormat.getOutputPath(jobconf));

        //int i = (job.waitForCompletion(true)? 1: 0);
View Full Code Here
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
public class InfiniDBOutputDriver extends Configured implements Tool
{
  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" };
    String [] outFields = {"id", "name"};
    jobconf.setInputFormat(IDBFileInputFormat.class);
    jobconf.setOutputFormat(InfiniDBOutputFormat.class);
    //Class.forName(jobconf.get(DBConfiguration.DRIVER_CLASS_PROPERTY));
    //Connection connection = DriverManager.getConnection("jdbc:mysql://srvswint4/tpch1?user=root");
   
        //Job job = new Job(jobconf, "infinidb_driver");
        //jobconf.setOutputKeyClass(InfiniDoopRecord.class);
    jobconf.setOutputKeyClass(NullWritable.class);
        jobconf.setOutputValueClass(Text.class);
        //jobconf.setMapOutputKeyClass(InfiniDoopRecord.class);
        //jobconf.setMapOutputValueClass(LongWritable.class);
       
        //InfiniDBInputFormat.setInput(jobconf, InfiniDoopRecord.class, "nation",
     //      null,  "n_nationkey", fields);
        InfiniDBOutputFormat.setOutput(jobconf, "zz1", outFields);
        //InfiniDBInputFormat.setInput(jobconf, InfiniDBRecord.class, "select * from nation", "select count(*) from nation");
    Job job = new Job(jobconf);
    jobconf.set("mapred.input.dir", "output1");
    jobconf.set("mapred.output.dir", "output");
   
        jobconf.setMapperClass(InfiniDoopMapper.class);
        jobconf.setNumMapTasks(1);
        jobconf.setNumReduceTasks(2);
        JobClient client = new JobClient();
        //System.out.println(jobconf.getUser());
        //System.out.println(FileOutputFormat.getOutputPath(jobconf));
        jobconf.setInt("mapred.min.split.size", 2147483647);

        //int i = (job.waitForCompletion(true)? 1: 0);
        //return i;
        client.setConf(jobconf);
    try {
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
   *                     job tracker.
   */
  public static void main(String[] args) throws IOException {
    Configuration defaults = new Configuration();
   
    JobConf conf = new JobConf(defaults, WordCount.class);
    conf.setJobName("wordcount");
    // the keys are words (strings)
    conf.setOutputKeyClass(UTF8.class);
    // the values are counts (ints)
    conf.setOutputValueClass(IntWritable.class);
   
    conf.setMapperClass(MapClass.class);       
    conf.setCombinerClass(Reduce.class);
    conf.setReducerClass(Reduce.class);
   
    List other_args = new ArrayList();
    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]);
        printUsage();
      } catch (ArrayIndexOutOfBoundsException except) {
        System.out.println("ERROR: Required parameter missing from " +
                           args[i-1]);
        printUsage(); // exits
      }
    }
    // 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.");
      printUsage();
    }
    conf.setInputDir(new File((String) other_args.get(0)));
    conf.setOutputDir(new File((String) other_args.get(1)));
   
    // Uncomment to run locally in a single process
    // conf.set("mapred.job.tracker", "local");
   
    JobClient.runJob(conf);
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
    File tempDir =
      new File("grep-temp-"+
               Integer.toString(new Random().nextInt(Integer.MAX_VALUE)));

    JobConf grepJob = new JobConf(defaults, Grep.class);
    grepJob.setJobName("grep-search");

    grepJob.setInputDir(new File(args[0]));

    grepJob.setMapperClass(RegexMapper.class);
    grepJob.set("mapred.mapper.regex", args[2]);
    if (args.length == 4)
      grepJob.set("mapred.mapper.regex.group", args[3]);
   
    grepJob.setCombinerClass(LongSumReducer.class);
    grepJob.setReducerClass(LongSumReducer.class);

    grepJob.setOutputDir(tempDir);
    grepJob.setOutputFormat(SequenceFileOutputFormat.class);
    grepJob.setOutputKeyClass(UTF8.class);
    grepJob.setOutputValueClass(LongWritable.class);

    JobClient.runJob(grepJob);

    JobConf sortJob = new JobConf(defaults, Grep.class);
    sortJob.setJobName("grep-sort");

    sortJob.setInputDir(tempDir);
    sortJob.setInputFormat(SequenceFileInputFormat.class);
    sortJob.setInputKeyClass(UTF8.class);
    sortJob.setInputValueClass(LongWritable.class);

    sortJob.setMapperClass(InverseMapper.class);

    sortJob.setNumReduceTasks(1);                 // write a single file
    sortJob.setOutputDir(new File(args[1]));
    sortJob.setOutputKeyComparatorClass           // sort by decreasing freq
      (LongWritable.DecreasingComparator.class);

    JobClient.runJob(sortJob);

    new JobClient(defaults).getFs().delete(tempDir);
View Full Code Here
170
171
172
173
174
175
176
177
178
179
  protected void setRequiredServiceHadoopConf(Configuration conf) {
    conf.set("fs.hdfs.impl.disable.cache", "true");
  }

  protected JobConf createHadoopConf(Configuration conf) {
    JobConf hadoopConf = new JobConf();
    XConfiguration.copy(serviceHadoopConf, hadoopConf);
    XConfiguration.copy(conf, hadoopConf);
    return hadoopConf;
  }
View Full Code Here
TOP

Related Classes of org.apache.hadoop.mapred.JobConf

  • bixo.config.BixoPlatform
  • com.cloudera.circus.test.TestXTest
  • com.cloudera.circus.test.XTest$TestMethodListener
  • com.cloudera.lib.service.hadoop.HadoopService
  • com.cloudera.lib.service.hadoop.TestHadoopService
  • com.facebook.hive.orc.TestInputOutputFormat
  • com.facebook.hiveio.output.HiveApiOutputFormat
  • com.facebook.hiveio.output.PerThread
  • com.facebook.LinkBench.LinkBenchDriverMR
  • com.facebook.presto.hive.HiveRecordSet
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.