Examples with PlanNodeStatsEstimate io.trino.cost.PlanNodeStatsEstimate used on opensource projects

Example 1 with PlanNodeStatsEstimate

use of io.trino.cost.PlanNodeStatsEstimate in project trino by trinodb.

the class DetermineSemiJoinDistributionType method canReplicate.

private boolean canReplicate(SemiJoinNode node, Context context) {
    DataSize joinMaxBroadcastTableSize = getJoinMaxBroadcastTableSize(context.getSession());
    PlanNode buildSide = node.getFilteringSource();
    PlanNodeStatsEstimate buildSideStatsEstimate = context.getStatsProvider().getStats(buildSide);
    double buildSideSizeInBytes = buildSideStatsEstimate.getOutputSizeInBytes(buildSide.getOutputSymbols(), context.getSymbolAllocator().getTypes());
    return buildSideSizeInBytes <= joinMaxBroadcastTableSize.toBytes() || getSourceTablesSizeInBytes(buildSide, context) <= joinMaxBroadcastTableSize.toBytes();
}

Example 2 with PlanNodeStatsEstimate

use of io.trino.cost.PlanNodeStatsEstimate in project trino by trinodb.

the class PushJoinIntoTableScan method getJoinStatistics.

private JoinStatistics getJoinStatistics(JoinNode join, TableScanNode left, TableScanNode right, Context context) {
    return new JoinStatistics() {

        @Override
        public Optional<BasicRelationStatistics> getLeftStatistics() {
            return getBasicRelationStats(left, left.getOutputSymbols(), context);
        }

        @Override
        public Optional<BasicRelationStatistics> getRightStatistics() {
            return getBasicRelationStats(right, right.getOutputSymbols(), context);
        }

        @Override
        public Optional<BasicRelationStatistics> getJoinStatistics() {
            return getBasicRelationStats(join, join.getOutputSymbols(), context);
        }

        private Optional<BasicRelationStatistics> getBasicRelationStats(PlanNode node, List<Symbol> outputSymbols, Context context) {
            PlanNodeStatsEstimate stats = context.getStatsProvider().getStats(node);
            TypeProvider types = context.getSymbolAllocator().getTypes();
            double outputRowCount = stats.getOutputRowCount();
            double outputSize = stats.getOutputSizeInBytes(outputSymbols, types);
            if (isNaN(outputRowCount) || isNaN(outputSize)) {
                return Optional.empty();
            }
            return Optional.of(new BasicRelationStatistics((long) outputRowCount, (long) outputSize));
        }
    };
}

Example 3 with PlanNodeStatsEstimate

use of io.trino.cost.PlanNodeStatsEstimate in project trino by trinodb.

the class TestReorderJoins method testReplicatesWhenNotRestricted.

@Test
public void testReplicatesWhenNotRestricted() {
    // variable width so that average row size is respected
    Type symbolType = createUnboundedVarcharType();
    int aRows = 10_000;
    int bRows = 10;
    PlanNodeStatsEstimate probeSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(aRows).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 640000, 10))).build();
    PlanNodeStatsEstimate buildSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(bRows).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000, 10))).build();
    // B table is small enough to be replicated in AUTOMATIC_RESTRICTED mode
    assertReorderJoins().setSystemProperty(JOIN_DISTRIBUTION_TYPE, AUTOMATIC.name()).setSystemProperty(JOIN_MAX_BROADCAST_TABLE_SIZE, "100MB").on(p -> {
        Symbol a1 = p.symbol("A1", symbolType);
        Symbol b1 = p.symbol("B1", symbolType);
        return p.join(INNER, p.values(new PlanNodeId("valuesA"), aRows, a1), p.values(new PlanNodeId("valuesB"), bRows, b1), ImmutableList.of(new EquiJoinClause(a1, b1)), ImmutableList.of(a1), ImmutableList.of(b1), Optional.empty());
    }).overrideStats("valuesA", probeSideStatsEstimate).overrideStats("valuesB", buildSideStatsEstimate).matches(join(INNER, ImmutableList.of(equiJoinClause("A1", "B1")), Optional.empty(), Optional.of(REPLICATED), values(ImmutableMap.of("A1", 0)), values(ImmutableMap.of("B1", 0))));
    probeSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(aRows).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 640000d * 10000, 10))).build();
    buildSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(bRows).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000d * 10000, 10))).build();
    // B table exceeds AUTOMATIC_RESTRICTED limit therefore it is partitioned
    assertReorderJoins().setSystemProperty(JOIN_DISTRIBUTION_TYPE, AUTOMATIC.name()).setSystemProperty(JOIN_MAX_BROADCAST_TABLE_SIZE, "100MB").on(p -> {
        Symbol a1 = p.symbol("A1", symbolType);
        Symbol b1 = p.symbol("B1", symbolType);
        return p.join(INNER, p.values(new PlanNodeId("valuesA"), aRows, a1), p.values(new PlanNodeId("valuesB"), bRows, b1), ImmutableList.of(new EquiJoinClause(a1, b1)), ImmutableList.of(a1), ImmutableList.of(b1), Optional.empty());
    }).overrideStats("valuesA", probeSideStatsEstimate).overrideStats("valuesB", buildSideStatsEstimate).matches(join(INNER, ImmutableList.of(equiJoinClause("A1", "B1")), Optional.empty(), Optional.of(PARTITIONED), values(ImmutableMap.of("A1", 0)), values(ImmutableMap.of("B1", 0))));
}

Example 4 with PlanNodeStatsEstimate

use of io.trino.cost.PlanNodeStatsEstimate in project trino by trinodb.

the class TestDetermineJoinDistributionType method testGetSourceTablesSizeInBytes.

@Test
public void testGetSourceTablesSizeInBytes() {
    PlanBuilder planBuilder = new PlanBuilder(new PlanNodeIdAllocator(), tester.getMetadata(), tester.getSession());
    Symbol symbol = planBuilder.symbol("col");
    Symbol sourceSymbol1 = planBuilder.symbol("source1");
    Symbol sourceSymbol2 = planBuilder.symbol("soruce2");
    // missing source stats
    assertEquals(getSourceTablesSizeInBytes(planBuilder.values(symbol), noLookup(), node -> PlanNodeStatsEstimate.unknown(), planBuilder.getTypes()), NaN);
    // two source plan nodes
    PlanNodeStatsEstimate sourceStatsEstimate1 = PlanNodeStatsEstimate.builder().setOutputRowCount(10).build();
    PlanNodeStatsEstimate sourceStatsEstimate2 = PlanNodeStatsEstimate.builder().setOutputRowCount(20).build();
    assertEquals(getSourceTablesSizeInBytes(planBuilder.union(ImmutableListMultimap.<Symbol, Symbol>builder().put(symbol, sourceSymbol1).put(symbol, sourceSymbol2).build(), ImmutableList.of(planBuilder.tableScan(ImmutableList.of(sourceSymbol1), ImmutableMap.of(sourceSymbol1, new TestingColumnHandle("col"))), planBuilder.values(new PlanNodeId("valuesNode"), sourceSymbol2))), noLookup(), node -> {
        if (node instanceof TableScanNode) {
            return sourceStatsEstimate1;
        }
        if (node instanceof ValuesNode) {
            return sourceStatsEstimate2;
        }
        return PlanNodeStatsEstimate.unknown();
    }, planBuilder.getTypes()), 270.0);
    // join node
    assertEquals(getSourceTablesSizeInBytes(planBuilder.join(INNER, planBuilder.values(sourceSymbol1), planBuilder.values(sourceSymbol2)), noLookup(), node -> sourceStatsEstimate1, planBuilder.getTypes()), NaN);
}

Example 5 with PlanNodeStatsEstimate

use of io.trino.cost.PlanNodeStatsEstimate in project trino by trinodb.

the class TestDetermineSemiJoinDistributionType method testReplicatesWhenNotRestricted.

@Test
public void testReplicatesWhenNotRestricted() {
    // variable width so that average row size is respected
    Type symbolType = createUnboundedVarcharType();
    int aRows = 10_000;
    int bRows = 10;
    PlanNodeStatsEstimate probeSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(aRows).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 640000, 10))).build();
    PlanNodeStatsEstimate buildSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(bRows).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000, 10))).build();
    // B table is small enough to be replicated in AUTOMATIC_RESTRICTED mode
    assertDetermineSemiJoinDistributionType().setSystemProperty(JOIN_DISTRIBUTION_TYPE, JoinDistributionType.AUTOMATIC.name()).setSystemProperty(JOIN_MAX_BROADCAST_TABLE_SIZE, "100MB").overrideStats("valuesA", probeSideStatsEstimate).overrideStats("valuesB", buildSideStatsEstimate).on(p -> {
        Symbol a1 = p.symbol("A1", symbolType);
        Symbol b1 = p.symbol("B1", symbolType);
        return p.semiJoin(p.values(new PlanNodeId("valuesA"), aRows, a1), p.values(new PlanNodeId("valuesB"), bRows, b1), a1, b1, p.symbol("output"), Optional.empty(), Optional.empty(), Optional.empty());
    }).matches(semiJoin("A1", "B1", "output", Optional.of(REPLICATED), values(ImmutableMap.of("A1", 0)), values(ImmutableMap.of("B1", 0))));
    probeSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(aRows).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 640000d * 10000, 10))).build();
    buildSideStatsEstimate = PlanNodeStatsEstimate.builder().setOutputRowCount(bRows).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000d * 10000, 10))).build();
    // B table exceeds AUTOMATIC_RESTRICTED limit therefore it is partitioned
    assertDetermineSemiJoinDistributionType().setSystemProperty(JOIN_DISTRIBUTION_TYPE, JoinDistributionType.AUTOMATIC.name()).setSystemProperty(JOIN_MAX_BROADCAST_TABLE_SIZE, "100MB").overrideStats("valuesA", probeSideStatsEstimate).overrideStats("valuesB", buildSideStatsEstimate).on(p -> {
        Symbol a1 = p.symbol("A1", symbolType);
        Symbol b1 = p.symbol("B1", symbolType);
        return p.semiJoin(p.values(new PlanNodeId("valuesA"), aRows, a1), p.values(new PlanNodeId("valuesB"), bRows, b1), a1, b1, p.symbol("output"), Optional.empty(), Optional.empty(), Optional.empty());
    }).matches(semiJoin("A1", "B1", "output", Optional.of(PARTITIONED), values(ImmutableMap.of("A1", 0)), values(ImmutableMap.of("B1", 0))));
}