Search in sources :

Example 36 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider in project asterixdb by apache.

the class CountOfCountsTest method countOfCountsExternalSortMultiNC.

@Test
public void countOfCountsExternalSortMultiNC() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] splits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "words.txt") };
    IFileSplitProvider splitProvider = new ConstantFileSplitProvider(splits);
    RecordDescriptor desc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor csvScanner = new FileScanOperatorDescriptor(spec, splitProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE }, ','), desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, csvScanner, NC2_ID);
    ExternalSortOperatorDescriptor sorter = new ExternalSortOperatorDescriptor(spec, 3, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, desc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    RecordDescriptor desc2 = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 0 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group, NC1_ID, NC2_ID, NC1_ID, NC2_ID);
    InMemorySortOperatorDescriptor sorter2 = new InMemorySortOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, desc2);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, sorter2, NC1_ID, NC2_ID);
    RecordDescriptor desc3 = new RecordDescriptor(new ISerializerDeserializer[] { IntegerSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
    PreclusteredGroupOperatorDescriptor group2 = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 1 }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(IntegerPointable.FACTORY) }, new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new CountFieldAggregatorFactory(true) }), desc3);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, group2, NC1_ID, NC2_ID);
    ResultSetId rsId = new ResultSetId(1);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, true, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    spec.addResultSetId(rsId);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor conn1 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn1, csvScanner, 0, sorter, 0);
    IConnectorDescriptor conn2 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn2, sorter, 0, group, 0);
    IConnectorDescriptor conn3 = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(conn3, group, 0, sorter2, 0);
    IConnectorDescriptor conn4 = new OneToOneConnectorDescriptor(spec);
    spec.connect(conn4, sorter2, 0, group2, 0);
    IConnectorDescriptor conn5 = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(conn5, group2, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) CountFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.CountFieldAggregatorFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileSplit(org.apache.hyracks.api.io.FileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) InMemorySortOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.InMemorySortOperatorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) ExternalSortOperatorDescriptor(org.apache.hyracks.dataflow.std.sort.ExternalSortOperatorDescriptor) PreclusteredGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.preclustered.PreclusteredGroupOperatorDescriptor) Test(org.junit.Test)

Example 37 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider in project asterixdb by apache.

the class Join method createJob.

private static JobSpecification createJob(FileSplit[] customerSplits, FileSplit[] orderSplits, FileSplit[] resultSplits, int numJoinPartitions, String algo, int graceInputSize, int graceRecordsPerFrame, double graceFactor, int memSize, int tableSize, boolean hasGroupBy, int frameSize) throws HyracksDataException {
    JobSpecification spec = new JobSpecification(frameSize);
    IFileSplitProvider custSplitsProvider = new ConstantFileSplitProvider(customerSplits);
    long custFileSize = 0;
    for (int i = 0; i < customerSplits.length; i++) {
        custFileSize += customerSplits[i].getFile(null).length();
    }
    IFileSplitProvider ordersSplitsProvider = new ConstantFileSplitProvider(orderSplits);
    long orderFileSize = 0;
    for (int i = 0; i < orderSplits.length; i++) {
        orderFileSize += orderSplits[i].getFile(null).length();
    }
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitsProvider, new DelimitedDataTupleParserFactory(orderParserFactories, '|'), Common.ordersDesc);
    createPartitionConstraint(spec, ordScanner, orderSplits);
    FileScanOperatorDescriptor custScanner = new FileScanOperatorDescriptor(spec, custSplitsProvider, new DelimitedDataTupleParserFactory(custParserFactories, '|'), Common.custDesc);
    createPartitionConstraint(spec, custScanner, customerSplits);
    IOperatorDescriptor join;
    if ("nestedloop".equalsIgnoreCase(algo)) {
        join = new NestedLoopJoinOperatorDescriptor(spec, new JoinComparatorFactory(PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), 0, 1), Common.custOrderJoinDesc, memSize, false, null);
    } else if ("inmem".equalsIgnoreCase(algo)) {
        join = new InMemoryHashJoinOperatorDescriptor(spec, new int[] { 0 }, new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, Common.custOrderJoinDesc, tableSize, null, memSize * frameSize);
    } else if ("hybrid".equalsIgnoreCase(algo)) {
        join = new OptimizedHybridHashJoinOperatorDescriptor(spec, memSize, graceInputSize, graceFactor, new int[] { 0 }, new int[] { 1 }, new IBinaryHashFunctionFamily[] { UTF8StringBinaryHashFunctionFamily.INSTANCE }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, Common.custOrderJoinDesc, new JoinComparatorFactory(PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), 0, 1), new JoinComparatorFactory(PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), 1, 0), null);
    } else {
        System.err.println("unknown algorithm:" + algo);
        return null;
    }
    PartitionConstraintHelper.addPartitionCountConstraint(spec, join, numJoinPartitions);
    IConnectorDescriptor ordJoinConn = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(ordJoinConn, ordScanner, 0, join, 1);
    IConnectorDescriptor custJoinConn = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
    spec.connect(custJoinConn, custScanner, 0, join, 0);
    IOperatorDescriptor endingOp = join;
    if (hasGroupBy) {
        RecordDescriptor groupResultDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), IntegerSerializerDeserializer.INSTANCE });
        ExternalGroupOperatorDescriptor gby = new ExternalGroupOperatorDescriptor(spec, tableSize, custFileSize + orderFileSize, new int[] { 6 }, memSize, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, new UTF8StringNormalizedKeyComputerFactory(), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, false), new IntSumFieldAggregatorFactory(3, false), new FloatSumFieldAggregatorFactory(5, false) }), new MultiFieldsAggregatorFactory(new IFieldAggregateDescriptorFactory[] { new IntSumFieldAggregatorFactory(1, false), new IntSumFieldAggregatorFactory(2, false), new FloatSumFieldAggregatorFactory(3, false) }), groupResultDesc, groupResultDesc, new HashSpillableTableFactory(new IBinaryHashFunctionFamily[] { UTF8StringBinaryHashFunctionFamily.INSTANCE }));
        createPartitionConstraint(spec, gby, resultSplits);
        IConnectorDescriptor joinGroupConn = new MToNPartitioningConnectorDescriptor(spec, new FieldHashPartitionComputerFactory(new int[] { 6 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }));
        spec.connect(joinGroupConn, join, 0, gby, 0);
        endingOp = gby;
    }
    IFileSplitProvider outSplitProvider = new ConstantFileSplitProvider(resultSplits);
    //FrameFileWriterOperatorDescriptor writer = new FrameFileWriterOperatorDescriptor(spec, outSplitProvider);
    IOperatorDescriptor writer = new PlainFileWriterOperatorDescriptor(spec, outSplitProvider, "|");
    createPartitionConstraint(spec, writer, resultSplits);
    IConnectorDescriptor endingPrinterConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(endingPrinterConn, endingOp, 0, writer, 0);
    spec.addRoot(writer);
    return spec;
}
Also used : OptimizedHybridHashJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.OptimizedHybridHashJoinOperatorDescriptor) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) HashSpillableTableFactory(org.apache.hyracks.dataflow.std.group.HashSpillableTableFactory) JoinComparatorFactory(org.apache.hyracks.dataflow.std.join.JoinComparatorFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) IBinaryHashFunctionFactory(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFactory) NestedLoopJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.NestedLoopJoinOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) IBinaryHashFunctionFamily(org.apache.hyracks.api.dataflow.value.IBinaryHashFunctionFamily) IFieldAggregateDescriptorFactory(org.apache.hyracks.dataflow.std.group.IFieldAggregateDescriptorFactory) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) UTF8StringNormalizedKeyComputerFactory(org.apache.hyracks.dataflow.common.data.normalizers.UTF8StringNormalizedKeyComputerFactory) MultiFieldsAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.MultiFieldsAggregatorFactory) ExternalGroupOperatorDescriptor(org.apache.hyracks.dataflow.std.group.external.ExternalGroupOperatorDescriptor) InMemoryHashJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.InMemoryHashJoinOperatorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) IBinaryComparatorFactory(org.apache.hyracks.api.dataflow.value.IBinaryComparatorFactory) MToNPartitioningConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNPartitioningConnectorDescriptor) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) FieldHashPartitionComputerFactory(org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory) Common.createPartitionConstraint(org.apache.hyracks.examples.tpch.client.Common.createPartitionConstraint) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) PlainFileWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.file.PlainFileWriterOperatorDescriptor) FloatSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.FloatSumFieldAggregatorFactory) IntSumFieldAggregatorFactory(org.apache.hyracks.dataflow.std.group.aggregators.IntSumFieldAggregatorFactory)

Example 38 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider in project asterixdb by apache.

the class TPCHCustomerOptimizedHybridHashJoinTest method customerOrderCIDHybridHashJoin_Case1.

@Test
public void customerOrderCIDHybridHashJoin_Case1() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] custSplits = new FileSplit[] { new ManagedFileSplit(NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer4.tbl") };
    IFileSplitProvider custSplitsProvider = new ConstantFileSplitProvider(custSplits);
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders4.tbl") };
    IFileSplitProvider ordersSplitsProvider = new ConstantFileSplitProvider(ordersSplits);
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitsProvider, new DelimitedDataTupleParserFactory(orderValueParserFactories, '|'), ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, ordScanner, NC2_ID);
    FileScanOperatorDescriptor custScanner = new FileScanOperatorDescriptor(spec, custSplitsProvider, new DelimitedDataTupleParserFactory(custValueParserFactories, '|'), custDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, custScanner, NC1_ID);
    OptimizedHybridHashJoinOperatorDescriptor join = new OptimizedHybridHashJoinOperatorDescriptor(spec, 15, 243, 1.2, new int[] { 0 }, new int[] { 1 }, new IBinaryHashFunctionFamily[] { UTF8StringBinaryHashFunctionFamily.INSTANCE }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, custOrderJoinDesc, new JoinComparatorFactory(PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), 0, 1), new JoinComparatorFactory(PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY), 1, 0), null);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, join, NC1_ID);
    String path = getClass().getName() + File.separator + "case1";
    IOperatorDescriptor printer = getPrinter(spec, path);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor custJoinConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(custJoinConn, custScanner, 0, join, 0);
    IConnectorDescriptor ordJoinConn = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(ordJoinConn, ordScanner, 0, join, 1);
    IConnectorDescriptor joinPrinterConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(joinPrinterConn, join, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
    System.out.println("output to " + path);
}
Also used : OptimizedHybridHashJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.OptimizedHybridHashJoinOperatorDescriptor) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) JoinComparatorFactory(org.apache.hyracks.dataflow.std.join.JoinComparatorFactory) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileSplit(org.apache.hyracks.api.io.FileSplit) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) Test(org.junit.Test)

Example 39 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider in project asterixdb by apache.

the class TPCHCustomerOrderHashJoinTest method customerOrderCIDJoin.

/*
     * TPCH Customer table: CREATE TABLE CUSTOMER ( C_CUSTKEY INTEGER NOT NULL,
     * C_NAME VARCHAR(25) NOT NULL, C_ADDRESS VARCHAR(40) NOT NULL, C_NATIONKEY
     * INTEGER NOT NULL, C_PHONE CHAR(15) NOT NULL, C_ACCTBAL DECIMAL(15,2) NOT
     * NULL, C_MKTSEGMENT CHAR(10) NOT NULL, C_COMMENT VARCHAR(117) NOT NULL );
     * TPCH Orders table: CREATE TABLE ORDERS ( O_ORDERKEY INTEGER NOT NULL,
     * O_CUSTKEY INTEGER NOT NULL, O_ORDERSTATUS CHAR(1) NOT NULL, O_TOTALPRICE
     * DECIMAL(15,2) NOT NULL, O_ORDERDATE DATE NOT NULL, O_ORDERPRIORITY
     * CHAR(15) NOT NULL, O_CLERK CHAR(15) NOT NULL, O_SHIPPRIORITY INTEGER NOT
     * NULL, O_COMMENT VARCHAR(79) NOT NULL );
     */
@Test
public void customerOrderCIDJoin() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] custSplits = new FileSplit[] { new ManagedFileSplit(NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer.tbl") };
    IFileSplitProvider custSplitsProvider = new ConstantFileSplitProvider(custSplits);
    RecordDescriptor custDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders.tbl") };
    IFileSplitProvider ordersSplitsProvider = new ConstantFileSplitProvider(ordersSplits);
    RecordDescriptor ordersDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    RecordDescriptor custOrderJoinDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitsProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, ordScanner, NC2_ID);
    FileScanOperatorDescriptor custScanner = new FileScanOperatorDescriptor(spec, custSplitsProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), custDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, custScanner, NC1_ID);
    InMemoryHashJoinOperatorDescriptor join = new InMemoryHashJoinOperatorDescriptor(spec, new int[] { 1 }, new int[] { 0 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, custOrderJoinDesc, 128, null, 128);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, join, NC1_ID);
    ResultSetId rsId = new ResultSetId(1);
    spec.addResultSetId(rsId);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, false, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor ordJoinConn = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(ordJoinConn, ordScanner, 0, join, 0);
    IConnectorDescriptor custJoinConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(custJoinConn, custScanner, 0, join, 1);
    IConnectorDescriptor joinPrinterConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(joinPrinterConn, join, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) InMemoryHashJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.InMemoryHashJoinOperatorDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileSplit(org.apache.hyracks.api.io.FileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) Test(org.junit.Test)

Example 40 with IFileSplitProvider

use of org.apache.hyracks.dataflow.std.file.IFileSplitProvider in project asterixdb by apache.

the class TPCHCustomerOrderHashJoinTest method customerOrderCIDHybridHashLeftOuterJoin.

@Test
public void customerOrderCIDHybridHashLeftOuterJoin() throws Exception {
    JobSpecification spec = new JobSpecification();
    FileSplit[] custSplits = new FileSplit[] { new ManagedFileSplit(NC1_ID, "data" + File.separator + "tpch0.001" + File.separator + "customer.tbl") };
    IFileSplitProvider custSplitsProvider = new ConstantFileSplitProvider(custSplits);
    RecordDescriptor custDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileSplit[] ordersSplits = new FileSplit[] { new ManagedFileSplit(NC2_ID, "data" + File.separator + "tpch0.001" + File.separator + "orders.tbl") };
    IFileSplitProvider ordersSplitsProvider = new ConstantFileSplitProvider(ordersSplits);
    RecordDescriptor ordersDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    RecordDescriptor custOrderJoinDesc = new RecordDescriptor(new ISerializerDeserializer[] { new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer(), new UTF8StringSerializerDeserializer() });
    FileScanOperatorDescriptor ordScanner = new FileScanOperatorDescriptor(spec, ordersSplitsProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), ordersDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, ordScanner, NC2_ID);
    FileScanOperatorDescriptor custScanner = new FileScanOperatorDescriptor(spec, custSplitsProvider, new DelimitedDataTupleParserFactory(new IValueParserFactory[] { UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE, UTF8StringParserFactory.INSTANCE }, '|'), custDesc);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, custScanner, NC1_ID);
    IMissingWriterFactory[] nonMatchWriterFactories = new IMissingWriterFactory[ordersDesc.getFieldCount()];
    for (int j = 0; j < nonMatchWriterFactories.length; j++) {
        nonMatchWriterFactories[j] = NoopMissingWriterFactory.INSTANCE;
    }
    HybridHashJoinOperatorDescriptor join = new HybridHashJoinOperatorDescriptor(spec, 32, 20, 200, 1.2, new int[] { 0 }, new int[] { 1 }, new IBinaryHashFunctionFactory[] { PointableBinaryHashFunctionFactory.of(UTF8StringPointable.FACTORY) }, new IBinaryComparatorFactory[] { PointableBinaryComparatorFactory.of(UTF8StringPointable.FACTORY) }, custOrderJoinDesc, null, true, nonMatchWriterFactories);
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, join, NC1_ID);
    ResultSetId rsId = new ResultSetId(1);
    spec.addResultSetId(rsId);
    IOperatorDescriptor printer = new ResultWriterOperatorDescriptor(spec, rsId, false, false, ResultSerializerFactoryProvider.INSTANCE.getResultSerializerFactoryProvider());
    PartitionConstraintHelper.addAbsoluteLocationConstraint(spec, printer, NC1_ID);
    IConnectorDescriptor ordJoinConn = new MToNBroadcastConnectorDescriptor(spec);
    spec.connect(ordJoinConn, ordScanner, 0, join, 1);
    IConnectorDescriptor custJoinConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(custJoinConn, custScanner, 0, join, 0);
    IConnectorDescriptor joinPrinterConn = new OneToOneConnectorDescriptor(spec);
    spec.connect(joinPrinterConn, join, 0, printer, 0);
    spec.addRoot(printer);
    runTest(spec);
}
Also used : IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) IValueParserFactory(org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory) IFileSplitProvider(org.apache.hyracks.dataflow.std.file.IFileSplitProvider) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) ConstantFileSplitProvider(org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider) IMissingWriterFactory(org.apache.hyracks.api.dataflow.value.IMissingWriterFactory) DelimitedDataTupleParserFactory(org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory) OneToOneConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) FileSplit(org.apache.hyracks.api.io.FileSplit) UTF8StringSerializerDeserializer(org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer) ResultWriterOperatorDescriptor(org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor) MToNBroadcastConnectorDescriptor(org.apache.hyracks.dataflow.std.connectors.MToNBroadcastConnectorDescriptor) ManagedFileSplit(org.apache.hyracks.api.io.ManagedFileSplit) ResultSetId(org.apache.hyracks.api.dataset.ResultSetId) IOperatorDescriptor(org.apache.hyracks.api.dataflow.IOperatorDescriptor) FileScanOperatorDescriptor(org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor) JobSpecification(org.apache.hyracks.api.job.JobSpecification) HybridHashJoinOperatorDescriptor(org.apache.hyracks.dataflow.std.join.HybridHashJoinOperatorDescriptor) Test(org.junit.Test)

Aggregations

IFileSplitProvider (org.apache.hyracks.dataflow.std.file.IFileSplitProvider)92 JobSpecification (org.apache.hyracks.api.job.JobSpecification)77 RecordDescriptor (org.apache.hyracks.api.dataflow.value.RecordDescriptor)64 ConstantFileSplitProvider (org.apache.hyracks.dataflow.std.file.ConstantFileSplitProvider)59 OneToOneConnectorDescriptor (org.apache.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor)58 IOperatorDescriptor (org.apache.hyracks.api.dataflow.IOperatorDescriptor)51 UTF8StringSerializerDeserializer (org.apache.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer)48 Test (org.junit.Test)45 FileScanOperatorDescriptor (org.apache.hyracks.dataflow.std.file.FileScanOperatorDescriptor)44 DelimitedDataTupleParserFactory (org.apache.hyracks.dataflow.std.file.DelimitedDataTupleParserFactory)43 FileSplit (org.apache.hyracks.api.io.FileSplit)39 ManagedFileSplit (org.apache.hyracks.api.io.ManagedFileSplit)39 IValueParserFactory (org.apache.hyracks.dataflow.common.data.parsers.IValueParserFactory)35 AlgebricksPartitionConstraint (org.apache.hyracks.algebricks.common.constraints.AlgebricksPartitionConstraint)30 IConnectorDescriptor (org.apache.hyracks.api.dataflow.IConnectorDescriptor)27 IIndexDataflowHelperFactory (org.apache.hyracks.storage.am.common.dataflow.IIndexDataflowHelperFactory)27 IndexDataflowHelperFactory (org.apache.hyracks.storage.am.common.dataflow.IndexDataflowHelperFactory)27 ResultSetId (org.apache.hyracks.api.dataset.ResultSetId)23 ResultWriterOperatorDescriptor (org.apache.hyracks.dataflow.std.result.ResultWriterOperatorDescriptor)22 FieldHashPartitionComputerFactory (org.apache.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory)21