Package eu.stratosphere.nephele.jobgraph

Examples of eu.stratosphere.nephele.jobgraph.JobInputVertex

220
221
222
223
224
225
226
227
228
229
230
231
232
233
  // -------------------------------------------------------------------------------------------------------------

  @SuppressWarnings("unchecked")
  private static JobInputVertex createPointsInput(JobGraph jobGraph, String pointsPath, int numSubTasks, TypeSerializerFactory<?> serializer) {
    CsvInputFormat pointsInFormat = new CsvInputFormat(' ', LongValue.class, LongValue.class, LongValue.class, LongValue.class);
    JobInputVertex pointsInput = JobGraphUtils.createInput(pointsInFormat, pointsPath, "Input[Points]", jobGraph, numSubTasks, numSubTasks);

    {
      TaskConfig taskConfig = new TaskConfig(pointsInput.getConfiguration());
      taskConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      taskConfig.setOutputSerializer(serializer);
    }

    return pointsInput;
View Full Code Here
234
235
236
237
238
239
240
241
242
243
244
245
246
247
  }

  @SuppressWarnings("unchecked")
  private static JobInputVertex createModelsInput(JobGraph jobGraph, String pointsPath, int numSubTasks, TypeSerializerFactory<?> serializer) {
    CsvInputFormat modelsInFormat = new CsvInputFormat(' ', LongValue.class, LongValue.class, LongValue.class, LongValue.class);
    JobInputVertex modelsInput = JobGraphUtils.createInput(modelsInFormat, pointsPath, "Input[Models]", jobGraph, numSubTasks, numSubTasks);

    {
      TaskConfig taskConfig = new TaskConfig(modelsInput.getConfiguration());
      taskConfig.addOutputShipStrategy(ShipStrategyType.BROADCAST);
      taskConfig.setOutputSerializer(serializer);
    }

    return modelsInput;
View Full Code Here
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
    final TypeSerializerFactory<?> serializer = RecordSerializerFactory.get();

    JobGraph jobGraph = new JobGraph("Distance Builder");

    // -- vertices ---------------------------------------------------------------------------------------------
    JobInputVertex points = createPointsInput(jobGraph, pointsPath, numSubTasks, serializer);
    JobInputVertex models = createModelsInput(jobGraph, centersPath, numSubTasks, serializer);
    JobTaskVertex mapper = createMapper(jobGraph, numSubTasks, serializer);
    JobOutputVertex output = createOutput(jobGraph, resultPath, numSubTasks, serializer);

    // -- edges ------------------------------------------------------------------------------------------------
    JobGraphUtils.connect(points, mapper, ChannelType.NETWORK, DistributionPattern.POINTWISE);
    JobGraphUtils.connect(models, mapper, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(mapper, output, ChannelType.NETWORK, DistributionPattern.POINTWISE);

    // -- instance sharing -------------------------------------------------------------------------------------
    points.setVertexToShareInstancesWith(output);
    models.setVertexToShareInstancesWith(output);
    mapper.setVertexToShareInstancesWith(output);

    return jobGraph;
  }
View Full Code Here
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
  private JobGraph createJobGraph(int dataVolumeGb, boolean useForwarder, boolean isSlowSender, boolean isSlowReceiver,
                  int numSubtasks, int numSubtasksPerInstance) throws JobGraphDefinitionException {

    JobGraph jobGraph = new JobGraph("Speed Test");

    JobInputVertex producer = new JobGenericInputVertex("Speed Test Producer", jobGraph);
    producer.setInputClass(SpeedTestProducer.class);
    producer.setNumberOfSubtasks(numSubtasks);
    producer.setNumberOfSubtasksPerInstance(numSubtasksPerInstance);
    producer.getConfiguration().setInteger(DATA_VOLUME_GB_CONFIG_KEY, dataVolumeGb);
    producer.getConfiguration().setBoolean(IS_SLOW_SENDER_CONFIG_KEY, isSlowSender);

    JobTaskVertex forwarder = null;
    if (useForwarder) {
      forwarder = new JobTaskVertex("Speed Test Forwarder", jobGraph);
      forwarder.setTaskClass(SpeedTestForwarder.class);
      forwarder.setNumberOfSubtasks(numSubtasks);
      forwarder.setNumberOfSubtasksPerInstance(numSubtasksPerInstance);
    }

    JobOutputVertex consumer = new JobOutputVertex("Speed Test Consumer", jobGraph);
    consumer.setOutputClass(SpeedTestConsumer.class);
    consumer.setNumberOfSubtasks(numSubtasks);
    consumer.setNumberOfSubtasksPerInstance(numSubtasksPerInstance);
    consumer.getConfiguration().setBoolean(IS_SLOW_RECEIVER_CONFIG_KEY, isSlowReceiver);

    if (useForwarder) {
      producer.connectTo(forwarder, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
      forwarder.connectTo(consumer, ChannelType.NETWORK, DistributionPattern.BIPARTITE);

      forwarder.setVertexToShareInstancesWith(producer);
      consumer.setVertexToShareInstancesWith(producer);
    }
    else {
      producer.connectTo(consumer, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
      producer.setVertexToShareInstancesWith(consumer);
    }

    return jobGraph;
  }
View Full Code Here
164
165
166
167
168
169
170
171
172
173
174
175
176
  private static JobInputVertex createVerticesInput(JobGraph jobGraph, String verticesPath, int numSubTasks,
      TypeSerializerFactory<?> serializer,
      TypeComparatorFactory<?> comparator) {
    @SuppressWarnings("unchecked")
    CsvInputFormat verticesInFormat = new CsvInputFormat(' ', LongValue.class);
    JobInputVertex verticesInput = JobGraphUtils.createInput(verticesInFormat, verticesPath, "VerticesInput",
      jobGraph, numSubTasks, numSubTasks);
    TaskConfig verticesInputConfig = new TaskConfig(verticesInput.getConfiguration());
    {
      verticesInputConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      verticesInputConfig.setOutputSerializer(serializer);

      // chained mapper that duplicates the id
View Full Code Here
196
197
198
199
200
201
202
203
204
205
206
207
208
      TypeSerializerFactory<?> serializer,
      TypeComparatorFactory<?> comparator) {
    // edges
    @SuppressWarnings("unchecked")
    CsvInputFormat edgesInFormat = new CsvInputFormat(' ', LongValue.class, LongValue.class);
    JobInputVertex edgesInput = JobGraphUtils.createInput(edgesInFormat, edgesPath, "EdgesInput", jobGraph,
      numSubTasks, numSubTasks);
    TaskConfig edgesInputConfig = new TaskConfig(edgesInput.getConfiguration());
    {
      edgesInputConfig.setOutputSerializer(serializer);
      edgesInputConfig.addOutputShipStrategy(ShipStrategyType.PARTITION_HASH);
      edgesInputConfig.setOutputComparator(comparator, 0);
    }
View Full Code Here
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
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
    final TypePairComparatorFactory<?, ?> pairComparator = RecordPairComparatorFactory.get();

    JobGraph jobGraph = new JobGraph("Connected Components (Unified Tails)");

    // -- invariant vertices -----------------------------------------------------------------------------------
    JobInputVertex vertices = createVerticesInput(jobGraph, verticesPath, numSubTasks, serializer, comparator);
    JobInputVertex edges = createEdgesInput(jobGraph, edgesPath, numSubTasks, serializer, comparator);
    JobTaskVertex head = createIterationHead(jobGraph, numSubTasks, serializer, comparator, pairComparator);

    JobTaskVertex intermediate = createIterationIntermediate(jobGraph, numSubTasks, serializer, comparator);
    TaskConfig intermediateConfig = new TaskConfig(intermediate.getConfiguration());

    JobOutputVertex output = createOutput(jobGraph, resultPath, numSubTasks, serializer);
    JobOutputVertex fakeTail = createFakeTail(jobGraph, numSubTasks);
    JobOutputVertex sync = createSync(jobGraph, numSubTasks, maxIterations);

    // --------------- the tail (solution set join) ---------------
    JobTaskVertex tail = JobGraphUtils.createTask(IterationTailPactTask.class, "IterationTail", jobGraph,
      numSubTasks, numSubTasks);
    TaskConfig tailConfig = new TaskConfig(tail.getConfiguration());
    {
      tailConfig.setIterationId(ITERATION_ID);

      tailConfig.setIsWorksetIteration();
      tailConfig.setIsWorksetUpdate();

      tailConfig.setIsSolutionSetUpdate();
      tailConfig.setIsSolutionSetUpdateWithoutReprobe();

      // inputs and driver
      tailConfig.addInputToGroup(0);
      tailConfig.setInputSerializer(serializer, 0);

      // output
      tailConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      tailConfig.setOutputSerializer(serializer);

      // the driver
      tailConfig.setDriver(JoinWithSolutionSetSecondDriver.class);
      tailConfig.setDriverStrategy(DriverStrategy.HYBRIDHASH_BUILD_SECOND);
      tailConfig.setDriverComparator(comparator, 0);
      tailConfig.setDriverPairComparator(pairComparator);
     
      tailConfig.setStubWrapper(new UserCodeClassWrapper<UpdateComponentIdMatch>(UpdateComponentIdMatch.class));
    }

    // -- edges ------------------------------------------------------------------------------------------------
    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(edges, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);

    JobGraphUtils.connect(head, intermediate, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    intermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, numSubTasks);

    JobGraphUtils.connect(intermediate, tail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    tailConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(head, output, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    JobGraphUtils.connect(tail, fakeTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(head, sync, ChannelType.NETWORK, DistributionPattern.POINTWISE);

    vertices.setVertexToShareInstancesWith(head);
    edges.setVertexToShareInstancesWith(head);

    intermediate.setVertexToShareInstancesWith(head);
    tail.setVertexToShareInstancesWith(head);

    output.setVertexToShareInstancesWith(head);
View Full Code Here
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
    final TypePairComparatorFactory<?, ?> pairComparator = RecordPairComparatorFactory.get();

    JobGraph jobGraph = new JobGraph("Connected Components (Unified Tails)");

    // input
    JobInputVertex vertices = createVerticesInput(jobGraph, verticesPath, numSubTasks, serializer, comparator);
    JobInputVertex edges = createEdgesInput(jobGraph, edgesPath, numSubTasks, serializer, comparator);

    // head
    JobTaskVertex head = createIterationHead(jobGraph, numSubTasks, serializer, comparator, pairComparator);
    TaskConfig headConfig = new TaskConfig(head.getConfiguration());
    headConfig.setWaitForSolutionSetUpdate();

    // intermediate
    JobTaskVertex intermediate = createIterationIntermediate(jobGraph, numSubTasks, serializer, comparator);
    TaskConfig intermediateConfig = new TaskConfig(intermediate.getConfiguration());

    // output and auxiliaries
    JobOutputVertex output = createOutput(jobGraph, resultPath, numSubTasks, serializer);
    JobOutputVertex ssFakeTail = createFakeTail(jobGraph, numSubTasks);
    JobOutputVertex wsFakeTail = createFakeTail(jobGraph, numSubTasks);
    JobOutputVertex sync = createSync(jobGraph, numSubTasks, maxIterations);

    // ------------------ the intermediate (ss join) ----------------------
    JobTaskVertex ssJoinIntermediate = JobGraphUtils.createTask(IterationIntermediatePactTask.class,
      "Solution Set Join", jobGraph, numSubTasks, numSubTasks);
    TaskConfig ssJoinIntermediateConfig = new TaskConfig(ssJoinIntermediate.getConfiguration());
    {
      ssJoinIntermediateConfig.setIterationId(ITERATION_ID);

      // inputs
      ssJoinIntermediateConfig.addInputToGroup(0);
      ssJoinIntermediateConfig.setInputSerializer(serializer, 0);

      // output
      ssJoinIntermediateConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      ssJoinIntermediateConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      ssJoinIntermediateConfig.setOutputComparator(comparator, 0);
      ssJoinIntermediateConfig.setOutputComparator(comparator, 1);

      ssJoinIntermediateConfig.setOutputSerializer(serializer);

      // driver
      ssJoinIntermediateConfig.setDriver(JoinWithSolutionSetSecondDriver.class);
      ssJoinIntermediateConfig.setDriverStrategy(DriverStrategy.HYBRIDHASH_BUILD_SECOND);
      ssJoinIntermediateConfig.setDriverComparator(comparator, 0);
      ssJoinIntermediateConfig.setDriverPairComparator(pairComparator);
     
      ssJoinIntermediateConfig.setStubWrapper(
        new UserCodeClassWrapper<UpdateComponentIdMatch>(UpdateComponentIdMatch.class));
    }

    // -------------------------- ss tail --------------------------------
    JobTaskVertex ssTail = JobGraphUtils.createTask(IterationTailPactTask.class, "IterationSolutionSetTail",
      jobGraph, numSubTasks, numSubTasks);
    TaskConfig ssTailConfig = new TaskConfig(ssTail.getConfiguration());
    {
      ssTailConfig.setIterationId(ITERATION_ID);
      ssTailConfig.setIsSolutionSetUpdate();
      ssTailConfig.setIsWorksetIteration();

      // inputs and driver
      ssTailConfig.addInputToGroup(0);
      ssTailConfig.setInputSerializer(serializer, 0);
      ssTailConfig.setInputAsynchronouslyMaterialized(0, true);
      ssTailConfig.setInputMaterializationMemory(0, MEM_PER_CONSUMER * JobGraphUtils.MEGABYTE);

      // output
      ssTailConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      ssTailConfig.setOutputSerializer(serializer);

      // the driver
      ssTailConfig.setDriver(CollectorMapDriver.class);
      ssTailConfig.setDriverStrategy(DriverStrategy.COLLECTOR_MAP);
      ssTailConfig.setStubWrapper(new UserCodeClassWrapper<DummyMapper>(DummyMapper.class));
    }

    // -------------------------- ws tail --------------------------------
    JobTaskVertex wsTail = JobGraphUtils.createTask(IterationTailPactTask.class, "IterationWorksetTail",
      jobGraph, numSubTasks, numSubTasks);
    TaskConfig wsTailConfig = new TaskConfig(wsTail.getConfiguration());
    {
      wsTailConfig.setIterationId(ITERATION_ID);
      wsTailConfig.setIsWorksetIteration();
      wsTailConfig.setIsWorksetUpdate();

      // inputs and driver
      wsTailConfig.addInputToGroup(0);
      wsTailConfig.setInputSerializer(serializer, 0);

      // output
      wsTailConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      wsTailConfig.setOutputSerializer(serializer);

      // the driver
      wsTailConfig.setDriver(CollectorMapDriver.class);
      wsTailConfig.setDriverStrategy(DriverStrategy.COLLECTOR_MAP);
      wsTailConfig.setStubWrapper(new UserCodeClassWrapper<DummyMapper>(DummyMapper.class));
    }

    // --------------- the wiring ---------------------

    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(edges, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);

    JobGraphUtils.connect(head, intermediate, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    intermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, numSubTasks);

    JobGraphUtils.connect(intermediate, ssJoinIntermediate, ChannelType.NETWORK, DistributionPattern.POINTWISE);
    ssJoinIntermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(ssJoinIntermediate, ssTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    ssTailConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(ssJoinIntermediate, wsTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    wsTailConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(head, output, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(ssTail, ssFakeTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    JobGraphUtils.connect(wsTail, wsFakeTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(head, sync, ChannelType.NETWORK, DistributionPattern.POINTWISE);

    vertices.setVertexToShareInstancesWith(head);
    edges.setVertexToShareInstancesWith(head);

    intermediate.setVertexToShareInstancesWith(head);

    ssJoinIntermediate.setVertexToShareInstancesWith(head);
    wsTail.setVertexToShareInstancesWith(head);
View Full Code Here
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
    final TypePairComparatorFactory<?, ?> pairComparator = RecordPairComparatorFactory.get();

    JobGraph jobGraph = new JobGraph("Connected Components (Intermediate Workset Update, Solution Set Tail)");

    // input
    JobInputVertex vertices = createVerticesInput(jobGraph, verticesPath, numSubTasks, serializer, comparator);
    JobInputVertex edges = createEdgesInput(jobGraph, edgesPath, numSubTasks, serializer, comparator);

    // head
    JobTaskVertex head = createIterationHead(jobGraph, numSubTasks, serializer, comparator, pairComparator);
    TaskConfig headConfig = new TaskConfig(head.getConfiguration());
    headConfig.setWaitForSolutionSetUpdate();

    // intermediate
    JobTaskVertex intermediate = createIterationIntermediate(jobGraph, numSubTasks, serializer, comparator);
    TaskConfig intermediateConfig = new TaskConfig(intermediate.getConfiguration());

    // output and auxiliaries
    JobOutputVertex output = createOutput(jobGraph, resultPath, numSubTasks, serializer);
    JobOutputVertex fakeTail = createFakeTail(jobGraph, numSubTasks);
    JobOutputVertex sync = createSync(jobGraph, numSubTasks, maxIterations);

    // ------------------ the intermediate (ws update) ----------------------
    JobTaskVertex wsUpdateIntermediate =
      JobGraphUtils.createTask(IterationIntermediatePactTask.class, "WorksetUpdate", jobGraph,
        numSubTasks, numSubTasks);
    TaskConfig wsUpdateConfig = new TaskConfig(wsUpdateIntermediate.getConfiguration());
    {
      wsUpdateConfig.setIterationId(ITERATION_ID);
      wsUpdateConfig.setIsWorksetIteration();
      wsUpdateConfig.setIsWorksetUpdate();

      // inputs
      wsUpdateConfig.addInputToGroup(0);
      wsUpdateConfig.setInputSerializer(serializer, 0);

      // output
      wsUpdateConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      wsUpdateConfig.setOutputComparator(comparator, 0);

      wsUpdateConfig.setOutputSerializer(serializer);

      // driver
      wsUpdateConfig.setDriver(JoinWithSolutionSetSecondDriver.class);
      wsUpdateConfig.setDriverStrategy(DriverStrategy.HYBRIDHASH_BUILD_SECOND);
      wsUpdateConfig.setDriverComparator(comparator, 0);
      wsUpdateConfig.setDriverPairComparator(pairComparator);
     
      wsUpdateConfig.setStubWrapper(new UserCodeClassWrapper<UpdateComponentIdMatch>(
        UpdateComponentIdMatch.class));
    }

    // -------------------------- ss tail --------------------------------
    JobTaskVertex ssTail =
      JobGraphUtils.createTask(IterationTailPactTask.class, "IterationSolutionSetTail", jobGraph,
        numSubTasks, numSubTasks);
    TaskConfig ssTailConfig = new TaskConfig(ssTail.getConfiguration());
    {
      ssTailConfig.setIterationId(ITERATION_ID);
      ssTailConfig.setIsSolutionSetUpdate();
      ssTailConfig.setIsWorksetIteration();

      // inputs and driver
      ssTailConfig.addInputToGroup(0);
      ssTailConfig.setInputSerializer(serializer, 0);

      // output
      ssTailConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      ssTailConfig.setOutputSerializer(serializer);

      // the driver
      ssTailConfig.setDriver(CollectorMapDriver.class);
      ssTailConfig.setDriverStrategy(DriverStrategy.COLLECTOR_MAP);
      ssTailConfig.setStubWrapper(new UserCodeClassWrapper<DummyMapper>(DummyMapper.class));
    }

    // edges

    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(edges, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);

    JobGraphUtils.connect(head, intermediate, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    intermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, numSubTasks);

    JobGraphUtils.connect(intermediate, wsUpdateIntermediate, ChannelType.NETWORK,
      DistributionPattern.POINTWISE);
    wsUpdateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(wsUpdateIntermediate, ssTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    ssTailConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(head, output, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(ssTail, fakeTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(head, sync, ChannelType.NETWORK, DistributionPattern.POINTWISE);

    vertices.setVertexToShareInstancesWith(head);
    edges.setVertexToShareInstancesWith(head);

    intermediate.setVertexToShareInstancesWith(head);

    wsUpdateIntermediate.setVertexToShareInstancesWith(head);
    ssTail.setVertexToShareInstancesWith(head);
View Full Code Here
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
    final TypePairComparatorFactory<?, ?> pairComparator = RecordPairComparatorFactory.get();

    JobGraph jobGraph = new JobGraph("Connected Components (Intermediate Solution Set Update, Workset Tail)");

    // input
    JobInputVertex vertices = createVerticesInput(jobGraph, verticesPath, numSubTasks, serializer, comparator);
    JobInputVertex edges = createEdgesInput(jobGraph, edgesPath, numSubTasks, serializer, comparator);

    // head
    JobTaskVertex head = createIterationHead(jobGraph, numSubTasks, serializer, comparator, pairComparator);

    // intermediate
    JobTaskVertex intermediate = createIterationIntermediate(jobGraph, numSubTasks, serializer, comparator);
    TaskConfig intermediateConfig = new TaskConfig(intermediate.getConfiguration());

    // output and auxiliaries
    JobOutputVertex output = createOutput(jobGraph, resultPath, numSubTasks, serializer);
    JobOutputVertex fakeTail = createFakeTail(jobGraph, numSubTasks);
    JobOutputVertex sync = createSync(jobGraph, numSubTasks, maxIterations);

    // ------------------ the intermediate (ss update) ----------------------
    JobTaskVertex ssJoinIntermediate = JobGraphUtils.createTask(IterationIntermediatePactTask.class,
      "Solution Set Update", jobGraph, numSubTasks, numSubTasks);
    TaskConfig ssJoinIntermediateConfig = new TaskConfig(ssJoinIntermediate.getConfiguration());
    {
      ssJoinIntermediateConfig.setIterationId(ITERATION_ID);
      ssJoinIntermediateConfig.setIsSolutionSetUpdate();
      ssJoinIntermediateConfig.setIsSolutionSetUpdateWithoutReprobe();

      // inputs
      ssJoinIntermediateConfig.addInputToGroup(0);
      ssJoinIntermediateConfig.setInputSerializer(serializer, 0);

      // output
      ssJoinIntermediateConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      ssJoinIntermediateConfig.setOutputComparator(comparator, 0);

      ssJoinIntermediateConfig.setOutputSerializer(serializer);

      // driver
      ssJoinIntermediateConfig.setDriver(JoinWithSolutionSetSecondDriver.class);
      ssJoinIntermediateConfig.setDriverStrategy(DriverStrategy.HYBRIDHASH_BUILD_SECOND);
      ssJoinIntermediateConfig.setDriverComparator(comparator, 0);
      ssJoinIntermediateConfig.setDriverPairComparator(pairComparator);
     
      ssJoinIntermediateConfig.setStubWrapper(new UserCodeClassWrapper<UpdateComponentIdMatch>(UpdateComponentIdMatch.class));
    }

    // -------------------------- ws tail --------------------------------
    JobTaskVertex wsTail = JobGraphUtils.createTask(IterationTailPactTask.class, "IterationWorksetTail",
      jobGraph, numSubTasks, numSubTasks);
    TaskConfig wsTailConfig = new TaskConfig(wsTail.getConfiguration());
    {
      wsTailConfig.setIterationId(ITERATION_ID);
      wsTailConfig.setIsWorksetIteration();
      wsTailConfig.setIsWorksetUpdate();

      // inputs and driver
      wsTailConfig.addInputToGroup(0);
      wsTailConfig.setInputSerializer(serializer, 0);

      // output
      wsTailConfig.addOutputShipStrategy(ShipStrategyType.FORWARD);
      wsTailConfig.setOutputSerializer(serializer);

      // the driver
      wsTailConfig.setDriver(CollectorMapDriver.class);
      wsTailConfig.setDriverStrategy(DriverStrategy.COLLECTOR_MAP);
      wsTailConfig.setStubWrapper(new UserCodeClassWrapper<DummyMapper>(DummyMapper.class));
    }

    // --------------- the wiring ---------------------

    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(edges, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    JobGraphUtils.connect(vertices, head, ChannelType.NETWORK, DistributionPattern.BIPARTITE);

    JobGraphUtils.connect(head, intermediate, ChannelType.NETWORK, DistributionPattern.BIPARTITE);
    intermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, numSubTasks);

    JobGraphUtils.connect(intermediate, ssJoinIntermediate, ChannelType.NETWORK, DistributionPattern.POINTWISE);
    ssJoinIntermediateConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(ssJoinIntermediate, wsTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);
    wsTailConfig.setGateIterativeWithNumberOfEventsUntilInterrupt(0, 1);

    JobGraphUtils.connect(head, output, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(wsTail, fakeTail, ChannelType.IN_MEMORY, DistributionPattern.POINTWISE);

    JobGraphUtils.connect(head, sync, ChannelType.NETWORK, DistributionPattern.POINTWISE);

    vertices.setVertexToShareInstancesWith(head);
    edges.setVertexToShareInstancesWith(head);

    intermediate.setVertexToShareInstancesWith(head);

    ssJoinIntermediate.setVertexToShareInstancesWith(head);
    wsTail.setVertexToShareInstancesWith(head);
View Full Code Here
TOP

Related Classes of eu.stratosphere.nephele.jobgraph.JobInputVertex

  • eu.stratosphere.compiler.plantranslate.NepheleJobGraphGenerator
  • eu.stratosphere.nephele.jobmanager.scheduler.queue.QueueSchedulerTest
  • eu.stratosphere.test.broadcastvars.BroadcastVarsNepheleITCase
  • eu.stratosphere.test.broadcastvars.KMeansIterativeNepheleITCase
  • eu.stratosphere.test.iterative.nephele.ConnectedComponentsNepheleITCase
  • eu.stratosphere.test.iterative.nephele.customdanglingpagerank.CustomCompensatableDanglingPageRank
  • eu.stratosphere.test.iterative.nephele.customdanglingpagerank.CustomCompensatableDanglingPageRankWithCombiner
  • eu.stratosphere.test.iterative.nephele.danglingpagerank.CompensatableDanglingPageRank
  • eu.stratosphere.test.iterative.nephele.IterationWithChainingNepheleITCase
  • eu.stratosphere.test.iterative.nephele.JobGraphUtils
    • 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.