Skip to content

HIVE-29365: Range predicate within the same histogram bucket leads to an estimate rowcount=1 #6319

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
thomasrebele wants to merge 1 commit into
base: master
Choose a base branch
from
+10 −5
Draft

HIVE-29365: Range predicate within the same histogram bucket leads to an estimate rowcount=1 #6319

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion ...rc/java/org/apache/hadoop/hive/ql/optimizer/calcite/stats/FilterSelectivityEstimator.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -492,7 +492,11 @@ public Double visitLiteral(RexLiteral literal) {
private static double rangedSelectivity(KllFloatsSketch kll, float val1, float val2) {
float[] splitPoints = new float[] { val1, val2 };
double[] boundaries = kll.getCDF(splitPoints, QuantileSearchCriteria.EXCLUSIVE);
return boundaries[1] - boundaries[0];
// due to the way the KLL sketch is constructed,
// it is not possible to differentiate selectivity values below the error
// (e.g., if the error is 2%, a real selectivity of 1.5% might be estimated as 0% by KLL)
double normalizedRankError = kll.getNormalizedRankError(false);
return Math.max(boundaries[1] - boundaries[0], normalizedRankError);
}

/**
Expand Down
9 changes: 5 additions & 4 deletions ...est/org/apache/hadoop/hive/ql/optimizer/calcite/stats/TestFilterSelectivityEstimator.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,6 +67,7 @@ public class TestFilterSelectivityEstimator {
private static final float[] VALUES = { 1, 2, 2, 2, 2, 2, 2, 2, 3, 4, 5, 6, 7 };
private static final KllFloatsSketch KLL = StatisticsTestUtils.createKll(VALUES);
private static final float DELTA = Float.MIN_VALUE;
private static final double MIN_KLL_SELECTIVITY = 0.013294757464848584;
private static final RexBuilder REX_BUILDER = new RexBuilder(new JavaTypeFactoryImpl(new HiveTypeSystemImpl()));
private static final RelDataTypeFactory TYPE_FACTORY = REX_BUILDER.getTypeFactory();
private static RelOptCluster relOptCluster;
Expand Down Expand Up @@ -240,12 +241,12 @@ public void testBetweenSelectivityLeftLowerThanMin() {

@Test
public void testBetweenSelectivityRightLowerThanMin() {
Assert.assertEquals(0, betweenSelectivity(KLL, -1, 0), DELTA);
Assert.assertEquals(MIN_KLL_SELECTIVITY, betweenSelectivity(KLL, -1, 0), DELTA);
}

@Test
public void testBetweenSelectivityLeftHigherThanMax() {
Assert.assertEquals(0, betweenSelectivity(KLL, 10, 11), DELTA);
Assert.assertEquals(MIN_KLL_SELECTIVITY, betweenSelectivity(KLL, 10, 11), DELTA);
}

@Test
Expand Down Expand Up @@ -399,15 +400,15 @@ public void testComputeRangePredicateSelectivityBetweenRightLowerThanMin() {
doReturn(Collections.singletonList(stats)).when(tableMock).getColStat(Collections.singletonList(0));
RexNode filter = REX_BUILDER.makeCall(HiveBetween.INSTANCE, boolFalse, inputRef0, intMinus1, int0);
FilterSelectivityEstimator estimator = new FilterSelectivityEstimator(scan, mq);
Assert.assertEquals(0, estimator.estimateSelectivity(filter), DELTA);
Assert.assertEquals(MIN_KLL_SELECTIVITY, estimator.estimateSelectivity(filter), DELTA);
}

@Test
public void testComputeRangePredicateSelectivityBetweenLeftHigherThanMax() {
doReturn(Collections.singletonList(stats)).when(tableMock).getColStat(Collections.singletonList(0));
RexNode filter = REX_BUILDER.makeCall(HiveBetween.INSTANCE, boolFalse, inputRef0, int10, int11);
FilterSelectivityEstimator estimator = new FilterSelectivityEstimator(scan, mq);
Assert.assertEquals(0, estimator.estimateSelectivity(filter), DELTA);
Assert.assertEquals(MIN_KLL_SELECTIVITY, estimator.estimateSelectivity(filter), DELTA);
}

@Test
Expand Down