Package org.apache.flink.api.java

Examples of org.apache.flink.api.java.ExecutionEnvironment

242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
    }
  }
 
  private Plan getJavaTestPlan(boolean joinPreservesSolutionSet, boolean mapBeforeSolutionDelta) {
   
    ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
    env.setDegreeOfParallelism(DEFAULT_PARALLELISM);
   
    @SuppressWarnings("unchecked")
    DataSet<Tuple3<Long, Long, Long>> solutionSetInput = env.fromElements(new Tuple3<Long, Long, Long>(1L, 2L, 3L)).name("Solution Set");
    @SuppressWarnings("unchecked")
    DataSet<Tuple3<Long, Long, Long>> worksetInput = env.fromElements(new Tuple3<Long, Long, Long>(1L, 2L, 3L)).name("Workset");
    @SuppressWarnings("unchecked")
    DataSet<Tuple3<Long, Long, Long>> invariantInput = env.fromElements(new Tuple3<Long, Long, Long>(1L, 2L, 3L)).name("Invariant Input");
   
    DeltaIteration<Tuple3<Long, Long, Long>, Tuple3<Long, Long, Long>> iter = solutionSetInput.iterateDelta(worksetInput, 100, 1, 2);
   
   
    DataSet<Tuple3<Long, Long, Long>> joinedWithSolutionSet =
   
    iter.getWorkset().join(invariantInput)
      .where(1, 2)
      .equalTo(1, 2)
      .with(new RichJoinFunction<Tuple3<Long,Long,Long>, Tuple3<Long, Long, Long>, Tuple3<Long,Long,Long>>() {
        public Tuple3<Long, Long, Long> join(Tuple3<Long, Long, Long> first, Tuple3<Long, Long, Long> second) {
          return first;
        }
      })
      .name(JOIN_WITH_INVARIANT_NAME)
   
    .join(iter.getSolutionSet())
      .where(1, 0)
      .equalTo(1, 2)
      .with(new RichJoinFunction<Tuple3<Long, Long, Long>, Tuple3<Long, Long, Long>, Tuple3<Long, Long, Long>>() {
        public Tuple3<Long, Long, Long> join(Tuple3<Long, Long, Long> first, Tuple3<Long, Long, Long> second) {
          return second;
        }
      })
      .name(JOIN_WITH_SOLUTION_SET)
      .withConstantSetSecond(joinPreservesSolutionSet ? new String[] {"0->0", "1->1", "2->2" } : null);
     
    DataSet<Tuple3<Long, Long, Long>> nextWorkset = joinedWithSolutionSet.groupBy(1, 2)
      .reduceGroup(new RichGroupReduceFunction<Tuple3<Long,Long,Long>, Tuple3<Long,Long,Long>>() {
        public void reduce(Iterable<Tuple3<Long, Long, Long>> values, Collector<Tuple3<Long, Long, Long>> out) {}
      })
      .name(NEXT_WORKSET_REDUCER_NAME)
      .withConstantSet("1->1","2->2","0->0");
   
   
    DataSet<Tuple3<Long, Long, Long>> nextSolutionSet = mapBeforeSolutionDelta ?
        joinedWithSolutionSet.map(new RichMapFunction<Tuple3<Long, Long, Long>,Tuple3<Long, Long, Long>>() { public Tuple3<Long, Long, Long> map(Tuple3<Long, Long, Long> value) { return value; } })
          .name(SOLUTION_DELTA_MAPPER_NAME).withConstantSet("0->0","1->1","2->2") :
        joinedWithSolutionSet;
   
    iter.closeWith(nextSolutionSet, nextWorkset)
      .print();
   
    return env.createProgramPlan();
  }
View Full Code Here
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
  }
 
  @Test
  public void testTwoIterationsWithMapperInbetween() throws Exception {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(8);
     
      DataSet<Tuple2<Long, Long>> verticesWithInitialId = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> edges = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> bulkResult = doBulkIteration(verticesWithInitialId, edges);
     
      DataSet<Tuple2<Long, Long>> mappedBulk = bulkResult.map(new DummyMap());
     
      DataSet<Tuple2<Long, Long>> depResult = doDeltaIteration(mappedBulk, edges);
     
      depResult.print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      assertEquals(1, op.getDataSinks().size());
      assertTrue(op.getDataSinks().iterator().next().getInput().getSource() instanceof WorksetIterationPlanNode);
     
View Full Code Here
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
  }
 
  @Test
  public void testTwoIterationsDirectlyChained() throws Exception {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(8);
     
      DataSet<Tuple2<Long, Long>> verticesWithInitialId = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> edges = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> bulkResult = doBulkIteration(verticesWithInitialId, edges);
     
      DataSet<Tuple2<Long, Long>> depResult = doDeltaIteration(bulkResult, edges);
     
      depResult.print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      assertEquals(1, op.getDataSinks().size());
      assertTrue(op.getDataSinks().iterator().next().getInput().getSource() instanceof WorksetIterationPlanNode);
     
View Full Code Here
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
  }
 
  @Test
  public void testTwoWorksetIterationsDirectlyChained() throws Exception {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(8);
     
      DataSet<Tuple2<Long, Long>> verticesWithInitialId = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> edges = env.fromElements(new Tuple2<Long, Long>(1L, 2L));
     
      DataSet<Tuple2<Long, Long>> firstResult = doDeltaIteration(verticesWithInitialId, edges);
     
      DataSet<Tuple2<Long, Long>> secondResult = doDeltaIteration(firstResult, edges);
     
      secondResult.print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      assertEquals(1, op.getDataSinks().size());
      assertTrue(op.getDataSinks().iterator().next().getInput().getSource() instanceof WorksetIterationPlanNode);
     
View Full Code Here
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
  }
 
  @Test
  public void testIterationPushingWorkOut() throws Exception {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(8);
     
      DataSet<Tuple2<Long, Long>> input1 = env.readCsvFile("/some/file/path").types(Long.class).map(new DuplicateValue());
     
      DataSet<Tuple2<Long, Long>> input2 = env.readCsvFile("/some/file/path").types(Long.class, Long.class);
     
      doBulkIteration(input1, input2).print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      assertEquals(1, op.getDataSinks().size());
      assertTrue(op.getDataSinks().iterator().next().getInput().getSource() instanceof BulkIterationPlanNode);
     
View Full Code Here
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 DeltaIterationDependenciesTest extends CompilerTestBase {

  @Test
  public void testExceptionWhenNewWorksetNotDependentOnWorkset() {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

      DataSet<Tuple2<Long, Long>> input = env.fromElements(new Tuple2<Long, Long>(0L, 0L));

      DeltaIteration<Tuple2<Long, Long>, Tuple2<Long, Long>> deltaIteration = input.iterateDelta(input, 10,0);

      DataSet<Tuple2<Long, Long>> delta = deltaIteration.getSolutionSet().join(deltaIteration.getWorkset())
                            .where(0).equalTo(0)
                            .projectFirst(1).projectSecond(1).types(Long.class, Long.class);

      DataSet<Tuple2<Long, Long>> nextWorkset = deltaIteration.getSolutionSet().join(input)
                            .where(0).equalTo(0)
                            .projectFirst(1).projectSecond(1).types(Long.class, Long.class);
     

      DataSet<Tuple2<Long, Long>> result = deltaIteration.closeWith(delta, nextWorkset);

      result.print();
     
      Plan p = env.createProgramPlan();
      try {
        compileNoStats(p);
        fail("Should not be able to compile, since the next workset does not depend on the workset");
      }
      catch (CompilerException e) {
View Full Code Here
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
public class IterationCompilerTest extends CompilerTestBase {

  @Test
  public void testIdentityIteration() {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(43);
     
      IterativeDataSet<Long> iteration = env.generateSequence(-4, 1000).iterate(100);
      iteration.closeWith(iteration).print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      new NepheleJobGraphGenerator().compileJobGraph(op);
    }
    catch (Exception e) {
View Full Code Here
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
  }
 
  @Test
  public void testEmptyWorksetIteration() {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(43);
     
      DataSet<Tuple2<Long, Long>> input = env.generateSequence(1, 20)
          .map(new MapFunction<Long, Tuple2<Long, Long>>() {
            @Override
            public Tuple2<Long, Long> map(Long value){ return null; }
          });
         
         
      DeltaIteration<Tuple2<Long, Long>, Tuple2<Long, Long>> iter = input.iterateDelta(input, 100, 0);
      iter.closeWith(iter.getWorkset(), iter.getWorkset())
        .print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      new NepheleJobGraphGenerator().compileJobGraph(op);
    }
    catch (Exception e) {
View Full Code Here
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
  }
 
  @Test
  public void testIterationWithUnionRoot() {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(43);
     
      IterativeDataSet<Long> iteration = env.generateSequence(-4, 1000).iterate(100);
     
      iteration.closeWith(
          iteration.map(new IdentityMapper<Long>()).union(iteration.map(new IdentityMapper<Long>())))
          .print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      new NepheleJobGraphGenerator().compileJobGraph(op);
    }
    catch (Exception e) {
View Full Code Here
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
  }
 
  @Test
  public void testWorksetIterationWithUnionRoot() {
    try {
      ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
      env.setDegreeOfParallelism(43);
     
      DataSet<Tuple2<Long, Long>> input = env.generateSequence(1, 20)
          .map(new MapFunction<Long, Tuple2<Long, Long>>() {
            @Override
            public Tuple2<Long, Long> map(Long value){ return null; }
          });
         
         
      DeltaIteration<Tuple2<Long, Long>, Tuple2<Long, Long>> iter = input.iterateDelta(input, 100, 0);
      iter.closeWith(
          iter.getWorkset().map(new IdentityMapper<Tuple2<Long,Long>>())
        .union(
          iter.getWorkset().map(new IdentityMapper<Tuple2<Long,Long>>()))
        , iter.getWorkset().map(new IdentityMapper<Tuple2<Long,Long>>())
        .union(
            iter.getWorkset().map(new IdentityMapper<Tuple2<Long,Long>>()))
        )
      .print();
     
      Plan p = env.createProgramPlan();
      OptimizedPlan op = compileNoStats(p);
     
      new NepheleJobGraphGenerator().compileJobGraph(op);
    }
    catch (Exception e) {
View Full Code Here
TOP

Related Classes of org.apache.flink.api.java.ExecutionEnvironment

  • org.apache.flink.api.avro.AvroOutputFormatTest
  • org.apache.flink.api.common.cache.DistributedCache.DistributedCacheEntry
  • org.apache.flink.api.common.operators.CollectionExecutionIterationTest
  • org.apache.flink.api.common.operators.CollectionExecutionWithBroadcastVariableTest
  • org.apache.flink.api.java.functions.SemanticPropertiesProjectionTest
  • org.apache.flink.api.java.functions.SemanticPropertiesTranslationTest
  • org.apache.flink.api.java.io.AvroInputFormatTypeExtractionTest
  • org.apache.flink.api.java.io.CsvReader
  • org.apache.flink.api.java.io.jdbc.example.JDBCExample
  • org.apache.flink.api.java.io.TextInputFormat
    • 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.