Examples of BulkIteration

Examples of eu.stratosphere.api.java.record.operators.BulkIteration

677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
  @Test
  public void testBranchBeforeIteration() {
    FileDataSource source1 = new FileDataSource(DummyInputFormat.class, IN_FILE, "Source 1");
    FileDataSource source2 = new FileDataSource(DummyInputFormat.class, IN_FILE, "Source 2");
   
    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(source2);
    iteration.setMaximumNumberOfIterations(10);
   
    MapOperator inMap = MapOperator.builder(new IdentityMap())
                               .input(source1)
                               .name("In Iteration Map")
                               .setBroadcastVariable("BC", iteration.getPartialSolution())
                               .build();
   
    iteration.setNextPartialSolution(inMap);
   
    MapOperator postMap = MapOperator.builder(new IdentityMap())
                     .input(source1)
                     .name("Post Iteration Map")
                     .setBroadcastVariable("BC", iteration)
View Full Code Here

Examples of eu.stratosphere.api.java.record.operators.BulkIteration

728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
    FileDataSource sourceB = new FileDataSource(DummyInputFormat.class, IN_FILE, "Source 2");

    FileDataSink sink1 = new FileDataSink(DummyOutputFormat.class, OUT_FILE, sourceA, "Sink 1");
    FileDataSink sink2 = new FileDataSink(DummyOutputFormat.class, OUT_FILE, sourceB, "Sink 2");

    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(sourceA);
    iteration.setMaximumNumberOfIterations(10);

    CrossOperator stepFunction = CrossOperator.builder(DummyCrossStub.class).name("StepFunction").
        input1(iteration.getPartialSolution()).
        input2(sourceB).
        build();

    iteration.setNextPartialSolution(stepFunction);

    FileDataSink sink3 = new FileDataSink(DummyOutputFormat.class, OUT_FILE, iteration, "Sink 3");

    List<FileDataSink> sinks = new ArrayList<FileDataSink>();
    sinks.add(sink1);
View Full Code Here

Examples of eu.stratosphere.api.java.record.operators.BulkIteration

906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
    ReduceOperator reducedSource = ReduceOperator.builder(IdentityReduce.class).
        input(source).
        name("Identity reduce source").
        build();

    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(mappedSource);
    iteration.setMaximumNumberOfIterations(10);

    JoinOperator nextPartialSolution = JoinOperator.builder(DummyMatchStub.class, IntValue.class, 0,0).
        input1(iteration.getPartialSolution()).
        input2(reducedSource).
        name("Next partial solution").
        build();

    iteration.setNextPartialSolution(nextPartialSolution);

    FileDataSink sink = new FileDataSink(DummyOutputFormat.class, OUT_FILE, iteration, "Iteration sink");
    List<FileDataSink> sinks = new ArrayList<FileDataSink>();
    sinks.add(sink);
View Full Code Here

Examples of eu.stratosphere.api.java.record.operators.BulkIteration

98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
    static Plan getTestPlan(int numSubTasks, String input, String output) {

        FileDataSource initialInput = new FileDataSource(new PointInFormat(), input, "Input");
        initialInput.setDegreeOfParallelism(1);

        BulkIteration iteration = new BulkIteration("Loop");
        iteration.setInput(initialInput);
        iteration.setMaximumNumberOfIterations(2);

        ReduceOperator dummyReduce = ReduceOperator.builder(new DummyReducer(), IntValue.class, 0)
                .input(iteration.getPartialSolution())
                .name("Reduce something")
                .build();


        MapOperator dummyMap = MapOperator.builder(new IdentityMapper()).input(dummyReduce).build();
        iteration.setNextPartialSolution(dummyMap);

        FileDataSink finalResult = new FileDataSink(new PointOutFormat(), output, iteration, "Output");

        Plan plan = new Plan(finalResult, "Iteration with chained map test");
        plan.setDefaultParallelism(numSubTasks);
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

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
   
    FileDataSource pageWithRankInput = new FileDataSource(new DanglingPageRankInputFormat(),
      pageWithRankInputPath, "DanglingPageWithRankInput");
    pageWithRankInput.getParameters().setLong(DanglingPageRankInputFormat.NUM_VERTICES_PARAMETER, numVertices);
   
    BulkIteration iteration = new BulkIteration("Page Rank Loop");
    iteration.setInput(pageWithRankInput);
   
    FileDataSource adjacencyListInput = new FileDataSource(new ImprovedAdjacencyListInputFormat(),
      adjacencyListInputPath, "AdjancencyListInput");
   
    JoinOperator join = JoinOperator.builder(new DotProductMatch(), LongValue.class, 0, 0)
        .input1(iteration.getPartialSolution())
        .input2(adjacencyListInput)
        .name("Join with Edges")
        .build();
   
    CoGroupOperator rankAggregation = CoGroupOperator.builder(new DotProductCoGroup(), LongValue.class, 0, 0)
        .input1(iteration.getPartialSolution())
        .input2(join)
        .name("Rank Aggregation")
        .build();
    rankAggregation.getParameters().setLong(DotProductCoGroup.NUM_VERTICES_PARAMETER, numVertices);
    rankAggregation.getParameters().setLong(DotProductCoGroup.NUM_DANGLING_VERTICES_PARAMETER, numDanglingVertices);
   
    iteration.setNextPartialSolution(rankAggregation);
    iteration.setMaximumNumberOfIterations(numIterations);
    iteration.getAggregators().registerAggregationConvergenceCriterion(DotProductCoGroup.AGGREGATOR_NAME, new PageRankStatsAggregator(),
        new DiffL1NormConvergenceCriterion());
   
    FileDataSink out = new FileDataSink(new PageWithRankOutFormat(), outputPath, iteration, "Final Ranks");

    Plan p = new Plan(out, "Dangling PageRank");
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
   
    FileDataSource pageWithRankInput = new FileDataSource(new DanglingPageRankInputFormat(),
      pageWithRankInputPath, "PageWithRank Input");
    pageWithRankInput.getParameters().setLong(NUM_VERTICES_CONFIG_PARAM, numVertices);
   
    BulkIteration iteration = new BulkIteration("Page Rank Loop");
    iteration.setInput(pageWithRankInput);
   
    FileDataSource adjacencyListInput = new FileDataSource(new ImprovedAdjacencyListInputFormat(),
      adjacencyListInputPath, "AdjancencyListInput");
   
    JoinOperator join = JoinOperator.builder(new JoinVerexWithEdgesMatch(), LongValue.class, 0, 0)
        .input1(iteration.getPartialSolution())
        .input2(adjacencyListInput)
        .name("Join with Edges")
        .build();
   
    ReduceOperator rankAggregation = ReduceOperator.builder(new AggregatingReduce(), LongValue.class, 0)
        .input(join)
        .name("Rank Aggregation")
        .build();
   
    iteration.setNextPartialSolution(rankAggregation);
    iteration.setMaximumNumberOfIterations(numIterations);
   
    JoinOperator termination = JoinOperator.builder(new JoinOldAndNew(), LongValue.class, 0, 0)
        .input1(iteration.getPartialSolution())
        .input2(rankAggregation)
        .name("Join Old and New")
        .build();
   
    iteration.setTerminationCriterion(termination);
   
    FileDataSink out = new FileDataSink(new PageWithRankOutFormat(), outputPath, iteration, "Final Ranks");

    Plan p = new Plan(out, "Simple PageRank");
    p.setDefaultParallelism(dop);
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
 
  static Plan getTestPlanPlan(int numSubTasks, String input, String output) {

    FileDataSource initialInput = new FileDataSource(TextInputFormat.class, input, "input");
   
    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(initialInput);
    iteration.setMaximumNumberOfIterations(NUM_ITERATIONS);

    ReduceOperator sumReduce = ReduceOperator.builder(new SumReducer())
        .input(iteration.getPartialSolution())
        .name("Compute sum (Reduce)")
        .build();
   
    iteration.setNextPartialSolution(sumReduce);

    @SuppressWarnings("unchecked")
    FileDataSink finalResult = new FileDataSink(new CsvOutputFormat("\n",  " ", StringValue.class), output, iteration, "Output");

    Plan plan = new Plan(finalResult, "Iteration with AllReducer (keyless Reducer)");
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

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
 
  private static Plan getTestPlanPlan(int numSubTasks, String input, String output) {

    FileDataSource initialInput = new FileDataSource(TextInputFormat.class, input, "input");
   
    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(initialInput);
    iteration.setMaximumNumberOfIterations(5);
    Assert.assertTrue(iteration.getMaximumNumberOfIterations() > 1);

    ReduceOperator sumReduce = ReduceOperator.builder(new SumReducer())
        .input(iteration.getPartialSolution())
        .name("Compute sum (Reduce)")
        .build();
   
    iteration.setNextPartialSolution(sumReduce);
   
    MapOperator terminationMapper = MapOperator.builder(new TerminationMapper())
        .input(iteration.getPartialSolution())
        .name("Compute termination criterion (Map)")
        .build();
   
    iteration.setTerminationCriterion(terminationMapper);

    FileDataSink finalResult = new FileDataSink(CsvOutputFormat.class, output, iteration, "Output");
    CsvOutputFormat.configureRecordFormat(finalResult)
      .recordDelimiter('\n')
      .fieldDelimiter(' ')
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
  static Plan getTestPlan(int numSubTasks, String input, String output) {

    FileDataSource initialInput = new FileDataSource(new PointInFormat(), input, "Input");
    initialInput.setDegreeOfParallelism(1);

    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(initialInput);
    iteration.setMaximumNumberOfIterations(2);

    ReduceOperator dummyReduce = ReduceOperator.builder(new DummyReducer(), IntValue.class, 0).input(iteration.getPartialSolution())
        .name("Reduce something").build();

    MapOperator dummyMap = MapOperator.builder(new IdentityMapper()).input(dummyReduce).build();
    iteration.setNextPartialSolution(dummyMap);

    FileDataSink finalResult = new FileDataSink(new PointOutFormat(), output, iteration, "Output");

    Plan plan = new Plan(finalResult, "Iteration with chained map test");
    plan.setDefaultParallelism(numSubTasks);
View Full Code Here

Examples of org.apache.flink.api.java.record.operators.BulkIteration

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
 
  private static Plan getTestPlanPlan(int numSubTasks, String input, String output) {

    FileDataSource initialInput = new FileDataSource(TextInputFormat.class, input, "input");
   
    BulkIteration iteration = new BulkIteration("Loop");
    iteration.setInput(initialInput);
    iteration.setMaximumNumberOfIterations(5);
    Assert.assertTrue(iteration.getMaximumNumberOfIterations() > 1);

    ReduceOperator sumReduce = ReduceOperator.builder(new SumReducer())
        .input(iteration.getPartialSolution())
        .name("Compute sum (Reduce)")
        .build();
   
    iteration.setNextPartialSolution(sumReduce);
   
    MapOperator terminationMapper = MapOperator.builder(new TerminationMapper())
        .input(sumReduce)
        .name("Compute termination criterion (Map)")
        .build();
   
    iteration.setTerminationCriterion(terminationMapper);

    FileDataSink finalResult = new FileDataSink(CsvOutputFormat.class, output, iteration, "Output");
    CsvOutputFormat.configureRecordFormat(finalResult)
      .recordDelimiter('\n')
      .fieldDelimiter(' ')
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.