use of io.prestosql.spi.plan.JoinNode.Type.INNER in project hetu-core by openlookeng.
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, 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, 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))));
}
use of io.prestosql.spi.plan.JoinNode.Type.INNER in project hetu-core by openlookeng.
the class TestReorderJoins method testReplicatesAndFlipsWhenOneTableMuchSmaller.
@Test
public void testReplicatesAndFlipsWhenOneTableMuchSmaller() {
// variable width so that average row size is respected
Type symbolType = createUnboundedVarcharType();
assertReorderJoins().on(p -> {
Symbol a1 = p.symbol("A1", symbolType);
Symbol b1 = p.symbol("B1", symbolType);
return p.join(INNER, p.values(new PlanNodeId("valuesA"), ImmutableList.of(a1), TWO_ROWS), p.values(new PlanNodeId("valuesB"), ImmutableList.of(b1), TWO_ROWS), ImmutableList.of(new EquiJoinClause(a1, b1)), ImmutableList.of(a1, b1), Optional.empty());
}).overrideStats("valuesA", PlanNodeStatsEstimate.builder().setOutputRowCount(100).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 6400, 100))).build()).overrideStats("valuesB", PlanNodeStatsEstimate.builder().setOutputRowCount(10000).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000, 100))).build()).matches(join(INNER, ImmutableList.of(equiJoinClause("B1", "A1")), Optional.empty(), Optional.of(REPLICATED), values(ImmutableMap.of("B1", 0)), values(ImmutableMap.of("A1", 0))));
}
use of io.prestosql.spi.plan.JoinNode.Type.INNER in project hetu-core by openlookeng.
the class TestJoinStatsRule method testStatsForInnerJoinWithTwoEquiClausesAndNonEqualityFunction.
@Test
public void testStatsForInnerJoinWithTwoEquiClausesAndNonEqualityFunction() {
double innerJoinRowCount = // driver join clause
LEFT_ROWS_COUNT * RIGHT_ROWS_COUNT / LEFT_JOIN_COLUMN_2_NDV * LEFT_JOIN_COLUMN_2_NON_NULLS * RIGHT_JOIN_COLUMN_2_NON_NULLS * // auxiliary join clause
UNKNOWN_FILTER_COEFFICIENT * // LEFT_JOIN_COLUMN < 10 non equality filter
0.3333333333;
PlanNodeStatsEstimate innerJoinStats = planNodeStats(innerJoinRowCount, symbolStatistics(LEFT_JOIN_COLUMN, 5.0, 10.0, 0.0, RIGHT_JOIN_COLUMN_NDV * 0.3333333333), symbolStatistics(RIGHT_JOIN_COLUMN, 5.0, 20.0, 0.0, RIGHT_JOIN_COLUMN_NDV), symbolStatistics(LEFT_JOIN_COLUMN_2, 100.0, 200.0, 0.0, RIGHT_JOIN_COLUMN_2_NDV), symbolStatistics(RIGHT_JOIN_COLUMN_2, 100.0, 200.0, 0.0, RIGHT_JOIN_COLUMN_2_NDV));
tester().assertStatsFor(pb -> {
Symbol leftJoinColumnSymbol = pb.symbol(LEFT_JOIN_COLUMN, BIGINT);
Symbol rightJoinColumnSymbol = pb.symbol(RIGHT_JOIN_COLUMN, DOUBLE);
Symbol leftJoinColumnSymbol2 = pb.symbol(LEFT_JOIN_COLUMN_2, BIGINT);
Symbol rightJoinColumnSymbol2 = pb.symbol(RIGHT_JOIN_COLUMN_2, DOUBLE);
ComparisonExpression leftJoinColumnLessThanTen = new ComparisonExpression(ComparisonExpression.Operator.LESS_THAN, toSymbolReference(leftJoinColumnSymbol), new LongLiteral("10"));
return pb.join(INNER, pb.values(leftJoinColumnSymbol, leftJoinColumnSymbol2), pb.values(rightJoinColumnSymbol, rightJoinColumnSymbol2), ImmutableList.of(new EquiJoinClause(leftJoinColumnSymbol2, rightJoinColumnSymbol2), new EquiJoinClause(leftJoinColumnSymbol, rightJoinColumnSymbol)), ImmutableList.of(leftJoinColumnSymbol, leftJoinColumnSymbol2, rightJoinColumnSymbol, rightJoinColumnSymbol2), Optional.of(castToRowExpression(leftJoinColumnLessThanTen)));
}).withSourceStats(0, planNodeStats(LEFT_ROWS_COUNT, LEFT_JOIN_COLUMN_STATS, LEFT_JOIN_COLUMN_2_STATS)).withSourceStats(1, planNodeStats(RIGHT_ROWS_COUNT, RIGHT_JOIN_COLUMN_STATS, RIGHT_JOIN_COLUMN_2_STATS)).check(stats -> stats.equalTo(innerJoinStats));
}
use of io.prestosql.spi.plan.JoinNode.Type.INNER in project hetu-core by openlookeng.
the class TransformFilteringSemiJoinToInnerJoin method apply.
@Override
public Result apply(FilterNode filterNode, Captures captures, Context context) {
SemiJoinNode semiJoin = captures.get(SEMI_JOIN);
// Do no transform semi-join in context of DELETE
if (PlanNodeSearcher.searchFrom(semiJoin.getSource(), context.getLookup()).where(node -> node instanceof TableScanNode && ((TableScanNode) node).isForDelete()).matches()) {
return Result.empty();
}
Symbol semiJoinSymbol = semiJoin.getSemiJoinOutput();
TypeProvider types = context.getSymbolAllocator().getTypes();
Predicate<RowExpression> isSemiJoinSymbol = expression -> expression.equals(toVariableReference(semiJoinSymbol, types));
LogicalRowExpressions logicalRowExpressions = new LogicalRowExpressions(new RowExpressionDeterminismEvaluator(metadata), new FunctionResolution(metadata.getFunctionAndTypeManager()), metadata.getFunctionAndTypeManager());
List<RowExpression> conjuncts = LogicalRowExpressions.extractConjuncts(filterNode.getPredicate());
if (conjuncts.stream().noneMatch(isSemiJoinSymbol)) {
return Result.empty();
}
RowExpression filteredPredicate = LogicalRowExpressions.and(conjuncts.stream().filter(expression -> !expression.equals(toVariableReference(semiJoinSymbol, types))).collect(toImmutableList()));
RowExpression simplifiedPredicate = inlineVariables(variable -> {
if (variable.equals(toVariableReference(semiJoinSymbol, types))) {
return TRUE_CONSTANT;
}
return variable;
}, filteredPredicate);
Optional<RowExpression> joinFilter = simplifiedPredicate.equals(TRUE_CONSTANT) ? Optional.empty() : Optional.of(simplifiedPredicate);
PlanNode filteringSourceDistinct = new AggregationNode(context.getIdAllocator().getNextId(), semiJoin.getFilteringSource(), ImmutableMap.of(), singleGroupingSet(ImmutableList.of(semiJoin.getFilteringSourceJoinSymbol())), ImmutableList.of(), SINGLE, Optional.empty(), Optional.empty(), AggregationNode.AggregationType.HASH, Optional.empty());
JoinNode innerJoin = new JoinNode(semiJoin.getId(), INNER, semiJoin.getSource(), filteringSourceDistinct, ImmutableList.of(new EquiJoinClause(semiJoin.getSourceJoinSymbol(), semiJoin.getFilteringSourceJoinSymbol())), semiJoin.getSource().getOutputSymbols(), joinFilter.isPresent() ? Optional.of(joinFilter.get()) : Optional.empty(), Optional.empty(), Optional.empty(), Optional.empty(), Optional.empty(), // TODO: dynamic filter from SemiJoinNode
ImmutableMap.of());
ProjectNode project = new ProjectNode(context.getIdAllocator().getNextId(), innerJoin, Assignments.builder().putAll(AssignmentUtils.identityAsSymbolReferences(innerJoin.getOutputSymbols())).put(semiJoinSymbol, TRUE_CONSTANT).build());
return Result.ofPlanNode(project);
}
use of io.prestosql.spi.plan.JoinNode.Type.INNER in project hetu-core by openlookeng.
the class TestDetermineJoinDistributionType method testFlipAndReplicateWhenOneTableMuchSmaller.
@Test
public void testFlipAndReplicateWhenOneTableMuchSmaller() {
// variable width so that average row size is respected
VarcharType symbolType = createUnboundedVarcharType();
int aRows = 100;
int bRows = 10_000;
assertDetermineJoinDistributionType().setSystemProperty(JOIN_DISTRIBUTION_TYPE, JoinDistributionType.AUTOMATIC.name()).overrideStats("valuesA", PlanNodeStatsEstimate.builder().setOutputRowCount(aRows).addSymbolStatistics(ImmutableMap.of(new Symbol("A1"), new SymbolStatsEstimate(0, 100, 0, 6400, 100))).build()).overrideStats("valuesB", PlanNodeStatsEstimate.builder().setOutputRowCount(bRows).addSymbolStatistics(ImmutableMap.of(new Symbol("B1"), new SymbolStatsEstimate(0, 100, 0, 640000, 100))).build()).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 JoinNode.EquiJoinClause(a1, b1)), ImmutableList.of(a1, b1), Optional.empty());
}).matches(join(INNER, ImmutableList.of(equiJoinClause("B1", "A1")), Optional.empty(), Optional.of(REPLICATED), values(ImmutableMap.of("B1", 0)), values(ImmutableMap.of("A1", 0))));
}
Aggregations