diff --git a/ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUtils.java b/ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUtils.java
index 65586278e2ec..dbb5565ff239 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUtils.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUtils.java
@@ -1578,9 +1578,11 @@ public static ColStatistics getColStatisticsFromExpression(HiveConf conf, Statis
             csList.add(cs);
           }
           if (csList.size() == engfd.getChildren().size()) {
-            Optional<ColStatistics> res = se.estimate(csList);
+            Optional<ColStatistics> res = se.estimate(csList, parentStats);
             if (res.isPresent()) {
               ColStatistics newStats = res.get();
+              // NDV cannot exceed numRows
+              newStats.setCountDistint(Math.min(newStats.getCountDistint(), numRows));
               colType = colType.toLowerCase();
               newStats.setColumnType(colType);
               newStats.setColumnName(colName);
@@ -2109,7 +2111,10 @@ private static List<Long> extractNDVGroupingColumns(List<ColStatistics> colStats
     for (ColStatistics cs : colStats) {
       if (cs != null) {
         long ndv = cs.getCountDistint();
-        if (cs.getNumNulls() > 0) {
+        // NDV needs to be adjusted if a column has a known NDV along with NULL values
+        // or if a column happens to be "const NULL"
+        if ((ndv > 0 && cs.getNumNulls() > 0) ||
+            (ndv == 0 && !cs.isEstimated() && cs.getNumNulls() == parentStats.getNumRows())) {
           ndv = StatsUtils.safeAdd(ndv, 1);
         }
         ndvValues.add(ndv);
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/PessimisticStatCombiner.java b/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/PessimisticStatCombiner.java
index 4de2867de7c0..48bb90820439 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/PessimisticStatCombiner.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/PessimisticStatCombiner.java
@@ -21,16 +21,26 @@
 import java.util.Optional;
 
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.stats.StatsUtils;
 
 /**
  * Combines {@link ColStatistics} objects to provide the most pessimistic estimate.
  */
 public class PessimisticStatCombiner {
 
+  private final long numRows;
   private boolean inited;
+  private boolean hasUnknownNDV;
   private ColStatistics result;
 
+  public PessimisticStatCombiner(long numRows) {
+    this.numRows = numRows;
+  }
+
   public void add(ColStatistics stat) {
+    // NDV==0 means unknown, unless it's a NULL constant (numNulls == numRows)
+    hasUnknownNDV = hasUnknownNDV || (stat.getCountDistint() == 0 && stat.getNumNulls() != numRows);
+
     if (!inited) {
       inited = true;
       result = stat.clone();
@@ -41,8 +51,10 @@ public void add(ColStatistics stat) {
     if (stat.getAvgColLen() > result.getAvgColLen()) {
       result.setAvgColLen(stat.getAvgColLen());
     }
-    if (stat.getCountDistint() > result.getCountDistint()) {
-      result.setCountDistint(stat.getCountDistint());
+    if (hasUnknownNDV) {
+      result.setCountDistint(0);
+    } else {
+      result.setCountDistint(StatsUtils.safeAdd(result.getCountDistint(), stat.getCountDistint()));
     }
     if (stat.getNumNulls() < 0 || result.getNumNulls() < 0) {
       result.setNumNulls(-1);
@@ -63,8 +75,8 @@ public void add(ColStatistics stat) {
       result.setFilterColumn();
     }
   }
+
   public Optional<ColStatistics> getResult() {
     return Optional.of(result);
-
   }
 }
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/StatEstimator.java b/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/StatEstimator.java
index 94aaa32ecfcb..80846fa24d30 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/StatEstimator.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/stats/estimator/StatEstimator.java
@@ -22,6 +22,7 @@
 import java.util.Optional;
 
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.plan.Statistics;
 
 /**
  * Enables statistics related computation on UDFs
@@ -39,5 +40,19 @@ public interface StatEstimator {
    * @param argStats the statistics for every argument of the UDF
    * @return {@link ColStatistics} estimate for the actual UDF.
    */
-  public Optional<ColStatistics> estimate(List<ColStatistics> argStats);
+  default Optional<ColStatistics> estimate(List<ColStatistics> argStats) {
+    throw new UnsupportedOperationException("This estimator requires parentStats");
+  }
+
+  /**
+   * Computes the output statistics with access to parent statistics.
+   * Override this method when the estimator uses more info for accurate estimation.
+   *
+   * @param argStats the statistics for every argument of the UDF
+   * @param parentStats statistics from the parent operator
+   * @return {@link ColStatistics} estimate for the actual UDF.
+   */
+  default Optional<ColStatistics> estimate(List<ColStatistics> argStats, Statistics parentStats) {
+    return estimate(argStats);
+  }
 }
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFCoalesce.java b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFCoalesce.java
index bbca9242ecaa..1799669bda57 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFCoalesce.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFCoalesce.java
@@ -26,6 +26,7 @@
 import org.apache.hadoop.hive.ql.exec.vector.VectorizedExpressionsSupportDecimal64;
 import org.apache.hadoop.hive.ql.metadata.HiveException;
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.plan.Statistics;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimator;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimatorProvider;
 import org.apache.hadoop.hive.ql.stats.estimator.PessimisticStatCombiner;
@@ -89,8 +90,8 @@ public StatEstimator getStatEstimator() {
   static class CoalesceStatEstimator implements StatEstimator {
 
     @Override
-    public Optional<ColStatistics> estimate(List<ColStatistics> argStats) {
-      PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    public Optional<ColStatistics> estimate(List<ColStatistics> argStats, Statistics parentStats) {
+      PessimisticStatCombiner combiner = new PessimisticStatCombiner(parentStats.getNumRows());
       for (int i = 0; i < argStats.size(); i++) {
         combiner.add(argStats.get(i));
       }
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFIf.java b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFIf.java
index eaa352317267..74bd2459debf 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFIf.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFIf.java
@@ -29,6 +29,7 @@
 import org.apache.hadoop.hive.ql.exec.vector.VectorizedExpressionsSupportDecimal64;
 import org.apache.hadoop.hive.ql.metadata.HiveException;
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.plan.Statistics;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimator;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimatorProvider;
 import org.apache.hadoop.hive.ql.stats.estimator.PessimisticStatCombiner;
@@ -187,8 +188,8 @@ public StatEstimator getStatEstimator() {
   static class IfStatEstimator implements StatEstimator {
 
     @Override
-    public Optional<ColStatistics> estimate(List<ColStatistics> argStats) {
-      PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    public Optional<ColStatistics> estimate(List<ColStatistics> argStats, Statistics parentStats) {
+      PessimisticStatCombiner combiner = new PessimisticStatCombiner(parentStats.getNumRows());
       combiner.add(argStats.get(1));
       combiner.add(argStats.get(2));
       return combiner.getResult();
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFWhen.java b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFWhen.java
index e6d3580692d3..5dab62ab959e 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFWhen.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFWhen.java
@@ -25,6 +25,7 @@
 import org.apache.hadoop.hive.ql.exec.UDFArgumentTypeException;
 import org.apache.hadoop.hive.ql.metadata.HiveException;
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.plan.Statistics;
 import org.apache.hadoop.hive.ql.stats.estimator.PessimisticStatCombiner;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimator;
 import org.apache.hadoop.hive.ql.stats.estimator.StatEstimatorProvider;
@@ -143,8 +144,8 @@ public StatEstimator getStatEstimator() {
   static class WhenStatEstimator implements StatEstimator {
 
     @Override
-    public Optional<ColStatistics> estimate(List<ColStatistics> argStats) {
-      PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    public Optional<ColStatistics> estimate(List<ColStatistics> argStats, Statistics parentStats) {
+      PessimisticStatCombiner combiner = new PessimisticStatCombiner(parentStats.getNumRows());
       for (int i = 1; i < argStats.size(); i += 2) {
         combiner.add(argStats.get(i));
       }
diff --git a/ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUtils.java b/ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUtils.java
index 3f76c554d446..48ee3b99cc35 100644
--- a/ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUtils.java
+++ b/ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUtils.java
@@ -25,6 +25,7 @@
 
 import java.lang.reflect.Field;
 import java.lang.reflect.Modifier;
+import java.util.Arrays;
 import java.util.Collections;
 import java.util.List;
 import java.util.Set;
@@ -39,7 +40,14 @@
 import org.apache.hadoop.hive.metastore.api.LongColumnStatsData;
 import org.apache.hadoop.hive.ql.plan.ColStatistics;
 import org.apache.hadoop.hive.ql.plan.ColStatistics.Range;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
 import org.apache.hadoop.hive.ql.plan.Statistics;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFCoalesce;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFIf;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFWhen;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
 import org.apache.hadoop.hive.serde.serdeConstants;
 import org.junit.jupiter.api.Test;
 import org.junit.jupiter.params.ParameterizedTest;
@@ -499,4 +507,223 @@ void testScaleColStatisticsPreservesUnknownNumFalses() {
     assertEquals(-1, colStats.get(0).getNumFalses(), "Unknown numFalses (-1) should be preserved after scaling");
   }
 
+  // Tests for computeNDVGroupingColumns / extractNDVGroupingColumns
+
+  @Test
+  void testComputeNDVGroupingColumnsSourceColumnWithNulls() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("col1", "string");
+    cs.setCountDistint(100);
+    cs.setNumNulls(50);
+    cs.setIsEstimated(false);  // source column
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(101, ndv, "Source column with nulls should get +1 for NULL: 100 + 1 = 101");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsSourceColumnNoNulls() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("col1", "string");
+    cs.setCountDistint(100);
+    cs.setNumNulls(0);
+    cs.setIsEstimated(false);
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(100, ndv, "Source column without nulls should not get +1");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsEstimatedExpression() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("case_expr", "string");
+    cs.setCountDistint(3);
+    cs.setNumNulls(500);
+    cs.setIsEstimated(true);  // computed expression (e.g., CASE)
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(4, ndv, "NDV with nulls: 3 + 1 = 4");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsAllNullColumn() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("col1", "string");
+    cs.setCountDistint(0);
+    cs.setNumNulls(1000);  // all rows are NULL
+    cs.setIsEstimated(false);
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(1, ndv, "All-NULL column: NDV=0 but numNulls==numRows, so NDV becomes 1");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsAllNullEstimatedColumn() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("case_expr", "string");
+    cs.setCountDistint(0);
+    cs.setNumNulls(1000);  // all rows are NULL
+    cs.setIsEstimated(true);  // from expression like CASE
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(0, ndv, "Estimated all-NULL column: NDV stays 0 (unknown from combiner)");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsUnknownNdv() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs = new ColStatistics("col1", "string");
+    cs.setCountDistint(0);  // unknown NDV
+    cs.setNumNulls(50);
+    cs.setIsEstimated(false);
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs), parentStats, false);
+    assertEquals(0, ndv, "Unknown NDV (0) should NOT get +1 to avoid false precision");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsMultipleColumns() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics cs1 = new ColStatistics("col1", "string");
+    cs1.setCountDistint(10);
+    cs1.setNumNulls(50);
+    cs1.setIsEstimated(false);
+
+    ColStatistics cs2 = new ColStatistics("col2", "int");
+    cs2.setCountDistint(5);
+    cs2.setNumNulls(0);
+    cs2.setIsEstimated(false);
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(cs1, cs2), parentStats, false);
+    // col1: 10 + 1 = 11 (has nulls), col2: 5 (no nulls)
+    // Product: 11 * 5 = 55
+    assertEquals(55, ndv, "Product of NDVs: (10+1) * 5 = 55");
+  }
+
+  @Test
+  void testComputeNDVGroupingColumnsMixedEstimatedAndSource() {
+    Statistics parentStats = new Statistics(1000, 8000, 0, 0);
+    parentStats.setColumnStatsState(Statistics.State.COMPLETE);
+
+    ColStatistics sourceCol = new ColStatistics("col1", "string");
+    sourceCol.setCountDistint(10);
+    sourceCol.setNumNulls(50);
+    sourceCol.setIsEstimated(false);  // source: gets +1
+
+    ColStatistics caseExpr = new ColStatistics("case_expr", "string");
+    caseExpr.setCountDistint(3);
+    caseExpr.setNumNulls(200);
+    caseExpr.setIsEstimated(true);  // estimated: no +1
+
+    long ndv = StatsUtils.computeNDVGroupingColumns(Arrays.asList(sourceCol, caseExpr), parentStats, false);
+    // sourceCol: 10 + 1 = 11, caseExpr: 3 + 1 = 4
+    // Product: 11 * 4 = 44
+    assertEquals(44, ndv, "Mixed columns: (10+1) * (3+1) = 44");
+  }
+
+  @Test
+  void testGetColStatisticsFromExpressionNdvCappedAtNumRows() {
+    Statistics parentStats = new Statistics(100, 800, 0, 0);
+
+    ColStatistics col1Stats = new ColStatistics("col1", "string");
+    col1Stats.setCountDistint(80);
+    col1Stats.setNumNulls(0);
+    col1Stats.setAvgColLen(10);
+
+    ColStatistics col2Stats = new ColStatistics("col2", "string");
+    col2Stats.setCountDistint(80);
+    col2Stats.setNumNulls(0);
+    col2Stats.setAvgColLen(10);
+
+    parentStats.setColumnStats(Arrays.asList(col1Stats, col2Stats));
+
+    GenericUDFIf udfIf = new GenericUDFIf();
+    ExprNodeConstantDesc condExpr = new ExprNodeConstantDesc(TypeInfoFactory.booleanTypeInfo, true);
+    ExprNodeColumnDesc col1Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col1", "t", false);
+    ExprNodeColumnDesc col2Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col2", "t", false);
+
+    ExprNodeGenericFuncDesc ifExpr = new ExprNodeGenericFuncDesc(
+        TypeInfoFactory.stringTypeInfo, udfIf, "if",
+        Arrays.asList(condExpr, col1Expr, col2Expr));
+
+    ColStatistics result = StatsUtils.getColStatisticsFromExpression(new HiveConf(), parentStats, ifExpr);
+
+    assertNotNull(result);
+    assertEquals(100, result.getCountDistint(), "NDV should be capped at numRows (100), not 160");
+  }
+
+  @Test
+  void testGetColStatisticsFromExpressionWhenNdvCapped() {
+    Statistics parentStats = new Statistics(100, 800, 0, 0);
+
+    ColStatistics col1Stats = new ColStatistics("col1", "string");
+    col1Stats.setCountDistint(60);
+    col1Stats.setNumNulls(0);
+    col1Stats.setAvgColLen(10);
+
+    ColStatistics col2Stats = new ColStatistics("col2", "string");
+    col2Stats.setCountDistint(70);
+    col2Stats.setNumNulls(0);
+    col2Stats.setAvgColLen(10);
+
+    parentStats.setColumnStats(Arrays.asList(col1Stats, col2Stats));
+
+    GenericUDFWhen udfWhen = new GenericUDFWhen();
+    ExprNodeConstantDesc condExpr = new ExprNodeConstantDesc(TypeInfoFactory.booleanTypeInfo, true);
+    ExprNodeColumnDesc col1Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col1", "t", false);
+    ExprNodeColumnDesc col2Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col2", "t", false);
+
+    ExprNodeGenericFuncDesc whenExpr = new ExprNodeGenericFuncDesc(
+        TypeInfoFactory.stringTypeInfo, udfWhen, "when",
+        Arrays.asList(condExpr, col1Expr, col2Expr));
+
+    ColStatistics result = StatsUtils.getColStatisticsFromExpression(new HiveConf(), parentStats, whenExpr);
+
+    assertNotNull(result);
+    assertEquals(100, result.getCountDistint(), "NDV should be capped at numRows (100), not 130");
+  }
+
+  @Test
+  void testGetColStatisticsFromExpressionCoalesceNdvCapped() {
+    Statistics parentStats = new Statistics(100, 800, 0, 0);
+
+    ColStatistics col1Stats = new ColStatistics("col1", "string");
+    col1Stats.setCountDistint(50);
+    col1Stats.setNumNulls(20);
+    col1Stats.setAvgColLen(10);
+
+    ColStatistics col2Stats = new ColStatistics("col2", "string");
+    col2Stats.setCountDistint(80);
+    col2Stats.setNumNulls(10);
+    col2Stats.setAvgColLen(10);
+
+    parentStats.setColumnStats(Arrays.asList(col1Stats, col2Stats));
+
+    GenericUDFCoalesce udfCoalesce = new GenericUDFCoalesce();
+    ExprNodeColumnDesc col1Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col1", "t", false);
+    ExprNodeColumnDesc col2Expr = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "col2", "t", false);
+
+    ExprNodeGenericFuncDesc coalesceExpr = new ExprNodeGenericFuncDesc(
+        TypeInfoFactory.stringTypeInfo, udfCoalesce, "coalesce",
+        Arrays.asList(col1Expr, col2Expr));
+
+    ColStatistics result = StatsUtils.getColStatisticsFromExpression(new HiveConf(), parentStats, coalesceExpr);
+
+    assertNotNull(result);
+    assertEquals(100, result.getCountDistint(), "NDV should be capped at numRows (100), not 130");
+  }
 }
diff --git a/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestPessimisticStatCombiner.java b/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestPessimisticStatCombiner.java
index 98bc589e40d3..281e7b82c27a 100644
--- a/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestPessimisticStatCombiner.java
+++ b/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestPessimisticStatCombiner.java
@@ -25,12 +25,51 @@
 
 class TestPessimisticStatCombiner {
 
+  @Test
+  void testNdvSumWhenBothKnown() {
+    ColStatistics stat1 = createStat("col1", "int", 50, 0, 4.0);
+    ColStatistics stat2 = createStat("col2", "int", 30, 0, 4.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
+    combiner.add(stat1);
+    combiner.add(stat2);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(80, result.getCountDistint(), "NDV should be summed: 50 + 30 = 80");
+  }
+
+  @Test
+  void testNdvUnknownPropagatedFromFirst() {
+    ColStatistics stat1 = createStat("col1", "int", 0, 0, 4.0);
+    ColStatistics stat2 = createStat("col2", "int", 100, 0, 4.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
+    combiner.add(stat1);
+    combiner.add(stat2);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(0, result.getCountDistint(), "Unknown NDV (0) should propagate");
+  }
+
+  @Test
+  void testNdvUnknownPropagatedFromSecond() {
+    ColStatistics stat1 = createStat("col1", "int", 100, 0, 4.0);
+    ColStatistics stat2 = createStat("col2", "int", 0, 0, 4.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
+    combiner.add(stat1);
+    combiner.add(stat2);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(0, result.getCountDistint(), "Unknown NDV (0) should propagate");
+  }
+
   @Test
   void testCombinePropagatesUnknownNumNullsFromFirst() {
     ColStatistics stat1 = createStat("col1", "int", 50, -1, 4.0); // unknown numNulls
     ColStatistics stat2 = createStat("col2", "int", 30, 100, 4.0);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -43,7 +82,7 @@ void testCombinePropagatesUnknownNumNullsFromSecond() {
     ColStatistics stat1 = createStat("col1", "int", 50, 100, 4.0);
     ColStatistics stat2 = createStat("col2", "int", 30, -1, 4.0); // unknown numNulls
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -61,7 +100,7 @@ void testCombinePropagatesUnknownNumTruesFromFirst() {
     stat2.setNumTrues(100);
     stat2.setNumFalses(150);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -79,7 +118,7 @@ void testCombinePropagatesUnknownNumTruesFromSecond() {
     stat2.setNumTrues(-1); // unknown
     stat2.setNumFalses(150);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -97,7 +136,7 @@ void testCombinePropagatesUnknownNumFalsesFromFirst() {
     stat2.setNumTrues(50);
     stat2.setNumFalses(150);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -115,7 +154,7 @@ void testCombinePropagatesUnknownNumFalsesFromSecond() {
     stat2.setNumTrues(50);
     stat2.setNumFalses(-1); // unknown
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -128,7 +167,7 @@ void testCombineBothUnknownNumNulls() {
     ColStatistics stat1 = createStat("col1", "int", 50, -1, 4.0);
     ColStatistics stat2 = createStat("col2", "int", 30, -1, 4.0);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -146,7 +185,7 @@ void testCombineBothUnknownNumTruesAndNumFalses() {
     stat2.setNumTrues(-1);
     stat2.setNumFalses(-1);
 
-    PessimisticStatCombiner combiner = new PessimisticStatCombiner();
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(1000);
     combiner.add(stat1);
     combiner.add(stat2);
 
@@ -155,6 +194,63 @@ void testCombineBothUnknownNumTruesAndNumFalses() {
     assertEquals(-1, combined.getNumFalses(), "Both unknown should result in unknown (-1)");
   }
 
+  @Test
+  void testNullConstantDoesNotContributeToNdv() {
+    long numRows = 100;
+    ColStatistics nullConstant = createStat("null", "int", 0, numRows, 0.0);
+    ColStatistics regularStat = createStat("col", "int", 50, 10, 4.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(numRows);
+    combiner.add(nullConstant);
+    combiner.add(regularStat);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(50, result.getCountDistint(), "NULL constant should not contribute to NDV");
+  }
+
+  @Test
+  void testNullConstantAsSecondDoesNotContributeToNdv() {
+    long numRows = 100;
+    ColStatistics regularStat = createStat("col", "int", 50, 10, 4.0);
+    ColStatistics nullConstant = createStat("null", "int", 0, numRows, 0.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(numRows);
+    combiner.add(regularStat);
+    combiner.add(nullConstant);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(50, result.getCountDistint(), "NULL constant should not contribute to NDV");
+  }
+
+  @Test
+  void testMultipleNullConstantsResultInZeroNdv() {
+    long numRows = 100;
+    ColStatistics nullConstant1 = createStat("null1", "int", 0, numRows, 0.0);
+    ColStatistics nullConstant2 = createStat("null2", "int", 0, numRows, 0.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(numRows);
+    combiner.add(nullConstant1);
+    combiner.add(nullConstant2);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(0, result.getCountDistint(), "Multiple NULL constants should result in NDV=0");
+    assertEquals(numRows, result.getNumNulls(), "numNulls should be numRows");
+  }
+
+  @Test
+  void testUnknownNdvNotConfusedWithNullConstant() {
+    long numRows = 100;
+    ColStatistics unknownNdv = createStat("col", "int", 0, 10, 4.0);
+    ColStatistics regularStat = createStat("col2", "int", 50, 5, 4.0);
+
+    PessimisticStatCombiner combiner = new PessimisticStatCombiner(numRows);
+    combiner.add(unknownNdv);
+    combiner.add(regularStat);
+
+    ColStatistics result = combiner.getResult().get();
+    assertEquals(0, result.getCountDistint(), "Unknown NDV should propagate as 0");
+  }
+
   private ColStatistics createStat(String name, String type, long ndv, long numNulls, double avgColLen) {
     ColStatistics stat = new ColStatistics(name, type);
     stat.setCountDistint(ndv);
diff --git a/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestStatEstimator.java b/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestStatEstimator.java
new file mode 100644
index 000000000000..7fd715f4a98d
--- /dev/null
+++ b/ql/src/test/org/apache/hadoop/hive/ql/stats/estimator/TestStatEstimator.java
@@ -0,0 +1,50 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hive.ql.stats.estimator;
+
+import static org.junit.jupiter.api.Assertions.assertThrows;
+
+import java.util.Arrays;
+import java.util.List;
+
+import org.apache.hadoop.hive.ql.plan.ColStatistics;
+import org.apache.hadoop.hive.ql.plan.Statistics;
+import org.junit.jupiter.api.Test;
+
+class TestStatEstimator {
+
+  @Test
+  void testDefaultEstimateThrowsUnsupportedOperation() {
+    StatEstimator estimator = new StatEstimator() {};
+    List<ColStatistics> argStats = Arrays.asList(new ColStatistics("col", "int"));
+
+    assertThrows(UnsupportedOperationException.class, () -> estimator.estimate(argStats),
+        "Default estimate(argStats) should throw UnsupportedOperationException");
+  }
+
+  @Test
+  void testDefaultEstimateWithParentStatsCallsEstimate() {
+    StatEstimator estimator = new StatEstimator() {};
+    List<ColStatistics> argStats = Arrays.asList(new ColStatistics("col", "int"));
+    Statistics parentStats = new Statistics(100, 800, 0, 0);
+
+    assertThrows(UnsupportedOperationException.class, () -> estimator.estimate(argStats, parentStats),
+        "Default estimate(argStats, parentStats) should delegate to estimate(argStats) which throws");
+  }
+}
diff --git a/ql/src/test/queries/clientpositive/ndv_case_const.q b/ql/src/test/queries/clientpositive/ndv_case_const.q
new file mode 100644
index 000000000000..7132e163aa13
--- /dev/null
+++ b/ql/src/test/queries/clientpositive/ndv_case_const.q
@@ -0,0 +1,40 @@
+-- Tests for CASE expression NDV estimation in Group By Operator.
+-- Verifies that "Statistics: Num rows" reflects accurate NDV computation
+-- when CASE branches contain constants, NULLs, and column references.
+CREATE TABLE t (cond INT, c2 STRING, c100 STRING);
+ALTER TABLE t UPDATE STATISTICS SET('numRows'='10000','rawDataSize'='1000000');
+ALTER TABLE t UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0');
+ALTER TABLE t UPDATE STATISTICS FOR COLUMN c2 SET('numDVs'='2','numNulls'='0','avgColLen'='5','maxColLen'='10');
+ALTER TABLE t UPDATE STATISTICS FOR COLUMN c100 SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10');
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE 'C' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE NULL END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' WHEN cond=3 THEN NULL ELSE 'B' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE c2 END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'C' ELSE c2 END x FROM t) sub GROUP BY x;
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE c100 END x FROM t) sub GROUP BY x;
+
+-- Test NDV cap: sum of branch NDVs (100+100+100+1=301) exceeds numRows (200)
+CREATE TABLE t_small (cond INT, c100a STRING, c100b STRING, c100c STRING);
+ALTER TABLE t_small UPDATE STATISTICS SET('numRows'='200','rawDataSize'='20000');
+ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0');
+ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100a SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10');
+ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100b SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10');
+ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100c SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10');
+
+EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c100a WHEN cond=2 THEN c100b WHEN cond=3 THEN c100c ELSE 'A' END x FROM t_small) sub GROUP BY x;
diff --git a/ql/src/test/results/clientpositive/llap/infer_bucket_sort_dyn_part.q.out b/ql/src/test/results/clientpositive/llap/infer_bucket_sort_dyn_part.q.out
index 995733564a08..17db16415c01 100644
--- a/ql/src/test/results/clientpositive/llap/infer_bucket_sort_dyn_part.q.out
+++ b/ql/src/test/results/clientpositive/llap/infer_bucket_sort_dyn_part.q.out
@@ -492,13 +492,13 @@ STAGE PLANS:
                       minReductionHashAggr: 0.99
                       mode: hash
                       outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                      Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                      Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                       Reduce Output Operator
                         key expressions: _col0 (type: string), _col1 (type: string)
                         null sort order: zz
                         sort order: ++
                         Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
         Reducer 3 
             Execution mode: vectorized, llap
@@ -508,14 +508,14 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     compressed: false
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                         output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/list_bucket_dml_6.q.out b/ql/src/test/results/clientpositive/llap/list_bucket_dml_6.q.out
index dbcf49b202e7..4f4a0b3df537 100644
--- a/ql/src/test/results/clientpositive/llap/list_bucket_dml_6.q.out
+++ b/ql/src/test/results/clientpositive/llap/list_bucket_dml_6.q.out
@@ -96,7 +96,7 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           bucketingVersion: 2
                           key expressions: _col0 (type: string), _col1 (type: string)
@@ -104,7 +104,7 @@ STAGE PLANS:
                           numBuckets: -1
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           tag: -1
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
                           auto parallelism: true
@@ -199,18 +199,18 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     bucketingVersion: 2
                     compressed: false
                     GlobalTableId: 0
 #### A masked pattern was here ####
                     NumFilesPerFileSink: 1
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
 #### A masked pattern was here ####
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
@@ -461,7 +461,7 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           bucketingVersion: 2
                           key expressions: _col0 (type: string), _col1 (type: string)
@@ -469,7 +469,7 @@ STAGE PLANS:
                           numBuckets: -1
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           tag: -1
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
                           auto parallelism: true
@@ -564,18 +564,18 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     bucketingVersion: 2
                     compressed: false
                     GlobalTableId: 0
 #### A masked pattern was here ####
                     NumFilesPerFileSink: 1
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
 #### A masked pattern was here ####
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
diff --git a/ql/src/test/results/clientpositive/llap/list_bucket_dml_7.q.out b/ql/src/test/results/clientpositive/llap/list_bucket_dml_7.q.out
index ad7051398156..6e45676ba107 100644
--- a/ql/src/test/results/clientpositive/llap/list_bucket_dml_7.q.out
+++ b/ql/src/test/results/clientpositive/llap/list_bucket_dml_7.q.out
@@ -96,7 +96,7 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           bucketingVersion: 2
                           key expressions: _col0 (type: string), _col1 (type: string)
@@ -104,7 +104,7 @@ STAGE PLANS:
                           numBuckets: -1
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           tag: -1
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
                           auto parallelism: true
@@ -199,18 +199,18 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     bucketingVersion: 2
                     compressed: false
                     GlobalTableId: 0
 #### A masked pattern was here ####
                     NumFilesPerFileSink: 1
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
 #### A masked pattern was here ####
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
@@ -461,7 +461,7 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           bucketingVersion: 2
                           key expressions: _col0 (type: string), _col1 (type: string)
@@ -469,7 +469,7 @@ STAGE PLANS:
                           numBuckets: -1
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           tag: -1
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
                           auto parallelism: true
@@ -564,18 +564,18 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     bucketingVersion: 2
                     compressed: false
                     GlobalTableId: 0
 #### A masked pattern was here ####
                     NumFilesPerFileSink: 1
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
 #### A masked pattern was here ####
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
diff --git a/ql/src/test/results/clientpositive/llap/list_bucket_dml_8.q.out b/ql/src/test/results/clientpositive/llap/list_bucket_dml_8.q.out
index 148303926d66..4e5651cccc53 100644
--- a/ql/src/test/results/clientpositive/llap/list_bucket_dml_8.q.out
+++ b/ql/src/test/results/clientpositive/llap/list_bucket_dml_8.q.out
@@ -96,7 +96,7 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           bucketingVersion: 2
                           key expressions: _col0 (type: string), _col1 (type: string)
@@ -104,7 +104,7 @@ STAGE PLANS:
                           numBuckets: -1
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           tag: -1
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
                           auto parallelism: true
@@ -199,18 +199,18 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     bucketingVersion: 2
                     compressed: false
                     GlobalTableId: 0
 #### A masked pattern was here ####
                     NumFilesPerFileSink: 1
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
 #### A masked pattern was here ####
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
diff --git a/ql/src/test/results/clientpositive/llap/merge_dynamic_partition4.q.out b/ql/src/test/results/clientpositive/llap/merge_dynamic_partition4.q.out
index 2c9c9015c173..85f1ea93c068 100644
--- a/ql/src/test/results/clientpositive/llap/merge_dynamic_partition4.q.out
+++ b/ql/src/test/results/clientpositive/llap/merge_dynamic_partition4.q.out
@@ -180,13 +180,13 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: string), _col1 (type: string)
                           null sort order: zz
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
             Execution mode: llap
             LLAP IO: no inputs
@@ -198,14 +198,14 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     compressed: false
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                         output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/merge_dynamic_partition5.q.out b/ql/src/test/results/clientpositive/llap/merge_dynamic_partition5.q.out
index 5b1e537b938a..ab9805c19485 100644
--- a/ql/src/test/results/clientpositive/llap/merge_dynamic_partition5.q.out
+++ b/ql/src/test/results/clientpositive/llap/merge_dynamic_partition5.q.out
@@ -156,13 +156,13 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                        Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: string), _col1 (type: string)
                           null sort order: zz
                           sort order: ++
                           Map-reduce partition columns: _col0 (type: string), _col1 (type: string)
-                          Statistics: Num rows: 1 Data size: 652 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 2 Data size: 1304 Basic stats: COMPLETE Column stats: COMPLETE
                           value expressions: _col2 (type: int), _col3 (type: struct<count:bigint,sum:double,input:int>), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: binary), _col7 (type: int), _col8 (type: struct<count:bigint,sum:double,input:int>), _col9 (type: bigint), _col10 (type: binary)
             Execution mode: llap
             LLAP IO: no inputs
@@ -174,14 +174,14 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10
-                Statistics: Num rows: 1 Data size: 516 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 2 Data size: 1032 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: 'STRING' (type: string), UDFToLong(COALESCE(_col2,0)) (type: bigint), COALESCE(_col3,0) (type: double), (_col4 - _col5) (type: bigint), COALESCE(ndv_compute_bit_vector(_col6),0) (type: bigint), _col6 (type: binary), 'STRING' (type: string), UDFToLong(COALESCE(_col7,0)) (type: bigint), COALESCE(_col8,0) (type: double), (_col4 - _col9) (type: bigint), COALESCE(ndv_compute_bit_vector(_col10),0) (type: bigint), _col10 (type: binary), _col0 (type: string), _col1 (type: string)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13
-                  Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                   File Output Operator
                     compressed: false
-                    Statistics: Num rows: 1 Data size: 712 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 2 Data size: 1424 Basic stats: COMPLETE Column stats: COMPLETE
                     table:
                         input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                         output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/ndv_case_const.q.out b/ql/src/test/results/clientpositive/llap/ndv_case_const.q.out
new file mode 100644
index 000000000000..a25b1e35b17d
--- /dev/null
+++ b/ql/src/test/results/clientpositive/llap/ndv_case_const.q.out
@@ -0,0 +1,868 @@
+PREHOOK: query: CREATE TABLE t (cond INT, c2 STRING, c100 STRING)
+PREHOOK: type: CREATETABLE
+PREHOOK: Output: database:default
+PREHOOK: Output: default@t
+POSTHOOK: query: CREATE TABLE t (cond INT, c2 STRING, c100 STRING)
+POSTHOOK: type: CREATETABLE
+POSTHOOK: Output: database:default
+POSTHOOK: Output: default@t
+PREHOOK: query: ALTER TABLE t UPDATE STATISTICS SET('numRows'='10000','rawDataSize'='1000000')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t
+PREHOOK: Output: default@t
+POSTHOOK: query: ALTER TABLE t UPDATE STATISTICS SET('numRows'='10000','rawDataSize'='1000000')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t
+POSTHOOK: Output: default@t
+PREHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t
+PREHOOK: Output: default@t
+POSTHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t
+POSTHOOK: Output: default@t
+PREHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN c2 SET('numDVs'='2','numNulls'='0','avgColLen'='5','maxColLen'='10')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t
+PREHOOK: Output: default@t
+POSTHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN c2 SET('numDVs'='2','numNulls'='0','avgColLen'='5','maxColLen'='10')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t
+POSTHOOK: Output: default@t
+PREHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN c100 SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t
+PREHOOK: Output: default@t
+POSTHOOK: query: ALTER TABLE t UPDATE STATISTICS FOR COLUMN c100 SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t
+POSTHOOK: Output: default@t
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE 'C' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE 'C' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN ('A') WHEN ((cond = 2)) THEN ('B') ELSE ('C') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 3 Data size: 255 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 3 Data size: 255 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 3 Data size: 255 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 3 Data size: 255 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN ('A') WHEN ((cond = 2)) THEN ('B') WHEN ((cond = 3)) THEN ('A') ELSE ('B') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 4 Data size: 340 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 4 Data size: 340 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 4 Data size: 340 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 4 Data size: 340 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE NULL END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE NULL END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN ('A') WHEN ((cond = 2)) THEN ('B') ELSE (null) END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN (null) WHEN ((cond = 2)) THEN ('A') ELSE ('B') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN 'A' ELSE 'B' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond) IN (1, 2)) THEN (null) WHEN ((cond = 3)) THEN ('A') ELSE ('B') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' WHEN cond=3 THEN NULL ELSE 'B' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN 'A' WHEN cond=3 THEN NULL ELSE 'B' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN (null) WHEN ((cond = 2)) THEN ('A') WHEN ((cond = 3)) THEN (null) ELSE ('B') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 40000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 3 Data size: 85 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN NULL WHEN cond=2 THEN NULL WHEN cond=3 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond) IN (1, 2)) THEN (null) WHEN ((cond = 3)) THEN (c100) ELSE ('A') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.9898
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 102 Data size: 89 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 102 Data size: 89 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 102 Data size: 89 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 102 Data size: 89 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE 'A' END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 1820000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN (c2) WHEN ((cond = 2)) THEN (c100) ELSE ('A') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 1820000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.9897
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 103 Data size: 9167 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 103 Data size: 9167 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 103 Data size: 9167 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 103 Data size: 9167 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE c2 END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c2 WHEN cond=2 THEN c100 ELSE c2 END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 1820000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN (c2) WHEN ((cond = 2)) THEN (c100) ELSE (c2) END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 1820000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.9896
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 104 Data size: 9256 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 104 Data size: 9256 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 104 Data size: 9256 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 104 Data size: 9256 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'C' ELSE c2 END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' WHEN cond=3 THEN 'C' ELSE c2 END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN ('A') WHEN ((cond = 2)) THEN ('B') WHEN ((cond = 3)) THEN ('C') ELSE (c2) END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.99
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 5 Data size: 445 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 5 Data size: 445 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 5 Data size: 445 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 5 Data size: 445 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE c100 END x FROM t) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN 'A' WHEN cond=2 THEN 'B' ELSE c100 END x FROM t) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t
+                  Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN ('A') WHEN ((cond = 2)) THEN ('B') ELSE (c100) END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 10000 Data size: 930000 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.9898
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 102 Data size: 9078 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 102 Data size: 9078 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 102 Data size: 9078 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 102 Data size: 9078 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
+PREHOOK: query: CREATE TABLE t_small (cond INT, c100a STRING, c100b STRING, c100c STRING)
+PREHOOK: type: CREATETABLE
+PREHOOK: Output: database:default
+PREHOOK: Output: default@t_small
+POSTHOOK: query: CREATE TABLE t_small (cond INT, c100a STRING, c100b STRING, c100c STRING)
+POSTHOOK: type: CREATETABLE
+POSTHOOK: Output: database:default
+POSTHOOK: Output: default@t_small
+PREHOOK: query: ALTER TABLE t_small UPDATE STATISTICS SET('numRows'='200','rawDataSize'='20000')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t_small
+PREHOOK: Output: default@t_small
+POSTHOOK: query: ALTER TABLE t_small UPDATE STATISTICS SET('numRows'='200','rawDataSize'='20000')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t_small
+POSTHOOK: Output: default@t_small
+PREHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t_small
+PREHOOK: Output: default@t_small
+POSTHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN cond SET('numDVs'='10','numNulls'='0')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t_small
+POSTHOOK: Output: default@t_small
+PREHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100a SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t_small
+PREHOOK: Output: default@t_small
+POSTHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100a SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t_small
+POSTHOOK: Output: default@t_small
+PREHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100b SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t_small
+PREHOOK: Output: default@t_small
+POSTHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100b SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t_small
+POSTHOOK: Output: default@t_small
+PREHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100c SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
+PREHOOK: Input: default@t_small
+PREHOOK: Output: default@t_small
+POSTHOOK: query: ALTER TABLE t_small UPDATE STATISTICS FOR COLUMN c100c SET('numDVs'='100','numNulls'='0','avgColLen'='5','maxColLen'='10')
+POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
+POSTHOOK: Input: default@t_small
+POSTHOOK: Output: default@t_small
+PREHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c100a WHEN cond=2 THEN c100b WHEN cond=3 THEN c100c ELSE 'A' END x FROM t_small) sub GROUP BY x
+PREHOOK: type: QUERY
+PREHOOK: Input: default@t_small
+#### A masked pattern was here ####
+POSTHOOK: query: EXPLAIN SELECT x FROM (SELECT CASE WHEN cond=1 THEN c100a WHEN cond=2 THEN c100b WHEN cond=3 THEN c100c ELSE 'A' END x FROM t_small) sub GROUP BY x
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@t_small
+#### A masked pattern was here ####
+STAGE DEPENDENCIES:
+  Stage-1 is a root stage
+  Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+  Stage: Stage-1
+    Tez
+#### A masked pattern was here ####
+      Edges:
+        Reducer 2 <- Map 1 (SIMPLE_EDGE)
+#### A masked pattern was here ####
+      Vertices:
+        Map 1 
+            Map Operator Tree:
+                TableScan
+                  alias: t_small
+                  Statistics: Num rows: 200 Data size: 54200 Basic stats: COMPLETE Column stats: COMPLETE
+                  Select Operator
+                    expressions: CASE WHEN ((cond = 1)) THEN (c100a) WHEN ((cond = 2)) THEN (c100b) WHEN ((cond = 3)) THEN (c100c) ELSE ('A') END (type: string)
+                    outputColumnNames: _col0
+                    Statistics: Num rows: 200 Data size: 54200 Basic stats: COMPLETE Column stats: COMPLETE
+                    Group By Operator
+                      keys: _col0 (type: string)
+                      minReductionHashAggr: 0.4
+                      mode: hash
+                      outputColumnNames: _col0
+                      Statistics: Num rows: 200 Data size: 17800 Basic stats: COMPLETE Column stats: COMPLETE
+                      Reduce Output Operator
+                        key expressions: _col0 (type: string)
+                        null sort order: z
+                        sort order: +
+                        Map-reduce partition columns: _col0 (type: string)
+                        Statistics: Num rows: 200 Data size: 17800 Basic stats: COMPLETE Column stats: COMPLETE
+            Execution mode: vectorized, llap
+            LLAP IO: all inputs
+        Reducer 2 
+            Execution mode: vectorized, llap
+            Reduce Operator Tree:
+              Group By Operator
+                keys: KEY._col0 (type: string)
+                mode: mergepartial
+                outputColumnNames: _col0
+                Statistics: Num rows: 200 Data size: 17800 Basic stats: COMPLETE Column stats: COMPLETE
+                File Output Operator
+                  compressed: false
+                  Statistics: Num rows: 200 Data size: 17800 Basic stats: COMPLETE Column stats: COMPLETE
+                  table:
+                      input format: org.apache.hadoop.mapred.SequenceFileInputFormat
+                      output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
+                      serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+      Processor Tree:
+        ListSink
+
diff --git a/ql/src/test/results/clientpositive/llap/parquet_vectorization_13.q.out b/ql/src/test/results/clientpositive/llap/parquet_vectorization_13.q.out
index 7125704c33d2..2a223991fe36 100644
--- a/ql/src/test/results/clientpositive/llap/parquet_vectorization_13.q.out
+++ b/ql/src/test/results/clientpositive/llap/parquet_vectorization_13.q.out
@@ -130,7 +130,7 @@ STAGE PLANS:
                               vectorProcessingMode: HASH
                               projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                           keys: _col0 (type: boolean), _col1 (type: tinyint), _col2 (type: timestamp), _col3 (type: float), _col4 (type: string)
-                          minReductionHashAggr: 0.4
+                          minReductionHashAggr: 0.99
                           mode: hash
                           outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14
                           Statistics: Num rows: 1386 Data size: 194258 Basic stats: COMPLETE Column stats: COMPLETE
@@ -487,7 +487,7 @@ STAGE PLANS:
                               vectorProcessingMode: HASH
                               projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                           keys: _col0 (type: boolean), _col1 (type: tinyint), _col2 (type: timestamp), _col3 (type: float), _col4 (type: string)
-                          minReductionHashAggr: 0.4
+                          minReductionHashAggr: 0.99
                           mode: hash
                           outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14
                           Statistics: Num rows: 1386 Data size: 194258 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/parquet_vectorization_14.q.out b/ql/src/test/results/clientpositive/llap/parquet_vectorization_14.q.out
index 5acc12c3b71d..12c88df0aebf 100644
--- a/ql/src/test/results/clientpositive/llap/parquet_vectorization_14.q.out
+++ b/ql/src/test/results/clientpositive/llap/parquet_vectorization_14.q.out
@@ -120,7 +120,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7]
                         keys: _col2 (type: string), _col1 (type: float), _col4 (type: double), _col0 (type: timestamp), _col3 (type: boolean)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12
                         Statistics: Num rows: 758 Data size: 130530 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/parquet_vectorization_15.q.out b/ql/src/test/results/clientpositive/llap/parquet_vectorization_15.q.out
index d7d3f4919183..3653c9466248 100644
--- a/ql/src/test/results/clientpositive/llap/parquet_vectorization_15.q.out
+++ b/ql/src/test/results/clientpositive/llap/parquet_vectorization_15.q.out
@@ -116,7 +116,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                         keys: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16
                         Statistics: Num rows: 6144 Data size: 1216372 Basic stats: COMPLETE Column stats: COMPLETE
@@ -154,16 +154,16 @@ STAGE PLANS:
                 keys: KEY._col0 (type: float), KEY._col1 (type: boolean), KEY._col2 (type: double), KEY._col3 (type: string), KEY._col4 (type: tinyint), KEY._col5 (type: int), KEY._col6 (type: timestamp)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16
-                Statistics: Num rows: 6144 Data size: 1216372 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 6104 Data size: 1208432 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp), power(((_col7 - ((_col8 * _col8) / _col9)) / if((_col9 = 1L), null, (_col9 - 1))), 0.5) (type: double), (-26.28 - CAST( _col5 AS decimal(10,0))) (type: decimal(13,2)), _col10 (type: double), (_col2 * 79.553D) (type: double), (33.0 % _col0) (type: float), power(((_col11 - ((_col12 * _col12) / _col13)) / if((_col13 = 1L), null, (_col13 - 1))), 0.5) (type: double), ((_col11 - ((_col12 * _col12) / _col13)) / _col13) (type: double), (-23.0D % _col2) (type: double), (- _col4) (type: tinyint), ((_col14 - ((_col15 * _col15) / _col16)) / if((_col16 = 1L), null, (_col16 - 1))) (type: double), (UDFToFloat(_col5) - _col0) (type: float), (-23 % UDFToInteger(_col4)) (type: int), (- (-26.28 - CAST( _col5 AS decimal(10,0)))) (type: decimal(13,2)), power(((_col14 - ((_col15 * _col15) / _col16)) / _col16), 0.5) (type: double)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20
-                  Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp)
                     null sort order: zzzzzzz
                     sort order: +++++++
-                    Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col7 (type: double), _col8 (type: decimal(13,2)), _col9 (type: double), _col10 (type: double), _col11 (type: float), _col12 (type: double), _col13 (type: double), _col14 (type: double), _col15 (type: tinyint), _col16 (type: double), _col17 (type: float), _col18 (type: int), _col19 (type: decimal(13,2)), _col20 (type: double)
         Reducer 3 
             Execution mode: llap
@@ -175,10 +175,10 @@ STAGE PLANS:
               Select Operator
                 expressions: KEY.reducesinkkey0 (type: float), KEY.reducesinkkey1 (type: boolean), KEY.reducesinkkey2 (type: double), KEY.reducesinkkey3 (type: string), KEY.reducesinkkey4 (type: tinyint), KEY.reducesinkkey5 (type: int), KEY.reducesinkkey6 (type: timestamp), VALUE._col0 (type: double), VALUE._col1 (type: decimal(13,2)), VALUE._col2 (type: double), VALUE._col3 (type: double), VALUE._col4 (type: float), VALUE._col5 (type: double), VALUE._col6 (type: double), VALUE._col7 (type: double), VALUE._col8 (type: tinyint), VALUE._col9 (type: double), VALUE._col10 (type: float), VALUE._col11 (type: int), VALUE._col12 (type: decimal(13,2)), VALUE._col13 (type: double)
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20
-                Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                 File Output Operator
                   compressed: false
-                  Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                   table:
                       input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                       output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/parquet_vectorization_16.q.out b/ql/src/test/results/clientpositive/llap/parquet_vectorization_16.q.out
index eeab9c89af72..a457b27af643 100644
--- a/ql/src/test/results/clientpositive/llap/parquet_vectorization_16.q.out
+++ b/ql/src/test/results/clientpositive/llap/parquet_vectorization_16.q.out
@@ -93,7 +93,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3]
                         keys: _col0 (type: string), _col1 (type: double), _col2 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6
                         Statistics: Num rows: 5979 Data size: 825318 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/parquet_vectorization_9.q.out b/ql/src/test/results/clientpositive/llap/parquet_vectorization_9.q.out
index eeab9c89af72..a457b27af643 100644
--- a/ql/src/test/results/clientpositive/llap/parquet_vectorization_9.q.out
+++ b/ql/src/test/results/clientpositive/llap/parquet_vectorization_9.q.out
@@ -93,7 +93,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3]
                         keys: _col0 (type: string), _col1 (type: double), _col2 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6
                         Statistics: Num rows: 5979 Data size: 825318 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/scratch_col_issue.q.out b/ql/src/test/results/clientpositive/llap/scratch_col_issue.q.out
index 5418ef860de3..1e23944fcaf7 100644
--- a/ql/src/test/results/clientpositive/llap/scratch_col_issue.q.out
+++ b/ql/src/test/results/clientpositive/llap/scratch_col_issue.q.out
@@ -189,7 +189,7 @@ STAGE PLANS:
                         outputColumnNames: _col1, _col2
                         input vertices:
                           1 Map 2
-                        Statistics: Num rows: 2 Data size: 368 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 1 Data size: 184 Basic stats: COMPLETE Column stats: COMPLETE
                         Select Operator
                           expressions: if((_col1) IN ('CertificateOfDeposit', 'RecurringDeposit', 'TermDeposit'), COALESCE(from_unixtime(to_unix_timestamp(CAST( _col2 AS DATE)), 'MM-dd-yyyy'),' '), '') (type: string)
                           outputColumnNames: _col0
@@ -198,13 +198,13 @@ STAGE PLANS:
                               native: true
                               projectedOutputColumnNums: [14]
                               selectExpressions: IfExprCondExprColumn(col 9:boolean, col 13:string, col 5:string)(children: StringColumnInList(col 1, values CertificateOfDeposit, RecurringDeposit, TermDeposit) -> 9:boolean, VectorCoalesce(columns [5, 12])(children: VectorUDFAdaptor(from_unixtime(to_unix_timestamp(CAST( _col2 AS DATE)), 'MM-dd-yyyy'))(children: VectorUDFUnixTimeStampDate(col 10)(children: CastStringToDate(col 2:string) -> 10:date) -> 11:bigint) -> 5:string, ConstantVectorExpression(val  ) -> 12:string) -> 13:string, ConstantVectorExpression(val ) -> 5:string) -> 14:string
-                          Statistics: Num rows: 2 Data size: 368 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 1 Data size: 184 Basic stats: COMPLETE Column stats: COMPLETE
                           File Output Operator
                             compressed: false
                             File Sink Vectorization:
                                 className: VectorFileSinkOperator
                                 native: false
-                            Statistics: Num rows: 2 Data size: 368 Basic stats: COMPLETE Column stats: COMPLETE
+                            Statistics: Num rows: 1 Data size: 184 Basic stats: COMPLETE Column stats: COMPLETE
                             table:
                                 input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                                 output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_13.q.out b/ql/src/test/results/clientpositive/llap/vectorization_13.q.out
index d1911fdb7f8b..0e96fffb2c09 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_13.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_13.q.out
@@ -131,7 +131,7 @@ STAGE PLANS:
                               vectorProcessingMode: HASH
                               projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                           keys: _col0 (type: boolean), _col1 (type: tinyint), _col2 (type: timestamp), _col3 (type: float), _col4 (type: string)
-                          minReductionHashAggr: 0.4
+                          minReductionHashAggr: 0.99
                           mode: hash
                           outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14
                           Statistics: Num rows: 1386 Data size: 194258 Basic stats: COMPLETE Column stats: COMPLETE
@@ -511,7 +511,7 @@ STAGE PLANS:
                               vectorProcessingMode: HASH
                               projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                           keys: _col0 (type: boolean), _col1 (type: tinyint), _col2 (type: timestamp), _col3 (type: float), _col4 (type: string)
-                          minReductionHashAggr: 0.4
+                          minReductionHashAggr: 0.99
                           mode: hash
                           outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14
                           Statistics: Num rows: 1386 Data size: 194258 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_14.q.out b/ql/src/test/results/clientpositive/llap/vectorization_14.q.out
index 25bfeb19bfcf..62feb4c66cff 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_14.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_14.q.out
@@ -121,7 +121,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7]
                         keys: _col2 (type: string), _col1 (type: float), _col4 (type: double), _col0 (type: timestamp), _col3 (type: boolean)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12
                         Statistics: Num rows: 758 Data size: 130530 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_15.q.out b/ql/src/test/results/clientpositive/llap/vectorization_15.q.out
index 6732aba7edd2..8c20b9f5fca7 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_15.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_15.q.out
@@ -117,7 +117,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
                         keys: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16
                         Statistics: Num rows: 6144 Data size: 1216372 Basic stats: COMPLETE Column stats: COMPLETE
@@ -163,16 +163,16 @@ STAGE PLANS:
                 keys: KEY._col0 (type: float), KEY._col1 (type: boolean), KEY._col2 (type: double), KEY._col3 (type: string), KEY._col4 (type: tinyint), KEY._col5 (type: int), KEY._col6 (type: timestamp)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16
-                Statistics: Num rows: 6144 Data size: 1216372 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 6104 Data size: 1208432 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp), power(((_col7 - ((_col8 * _col8) / _col9)) / if((_col9 = 1L), null, (_col9 - 1))), 0.5) (type: double), (-26.28 - CAST( _col5 AS decimal(10,0))) (type: decimal(13,2)), _col10 (type: double), (_col2 * 79.553D) (type: double), (33.0 % _col0) (type: float), power(((_col11 - ((_col12 * _col12) / _col13)) / if((_col13 = 1L), null, (_col13 - 1))), 0.5) (type: double), ((_col11 - ((_col12 * _col12) / _col13)) / _col13) (type: double), (-23.0D % _col2) (type: double), (- _col4) (type: tinyint), ((_col14 - ((_col15 * _col15) / _col16)) / if((_col16 = 1L), null, (_col16 - 1))) (type: double), (UDFToFloat(_col5) - _col0) (type: float), (-23 % UDFToInteger(_col4)) (type: int), (- (-26.28 - CAST( _col5 AS decimal(10,0)))) (type: decimal(13,2)), power(((_col14 - ((_col15 * _col15) / _col16)) / _col16), 0.5) (type: double)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20
-                  Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: float), _col1 (type: boolean), _col2 (type: double), _col3 (type: string), _col4 (type: tinyint), _col5 (type: int), _col6 (type: timestamp)
                     null sort order: zzzzzzz
                     sort order: +++++++
-                    Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col7 (type: double), _col8 (type: decimal(13,2)), _col9 (type: double), _col10 (type: double), _col11 (type: float), _col12 (type: double), _col13 (type: double), _col14 (type: double), _col15 (type: tinyint), _col16 (type: double), _col17 (type: float), _col18 (type: int), _col19 (type: decimal(13,2)), _col20 (type: double)
         Reducer 3 
             Execution mode: llap
@@ -184,10 +184,10 @@ STAGE PLANS:
               Select Operator
                 expressions: KEY.reducesinkkey0 (type: float), KEY.reducesinkkey1 (type: boolean), KEY.reducesinkkey2 (type: double), KEY.reducesinkkey3 (type: string), KEY.reducesinkkey4 (type: tinyint), KEY.reducesinkkey5 (type: int), KEY.reducesinkkey6 (type: timestamp), VALUE._col0 (type: double), VALUE._col1 (type: decimal(13,2)), VALUE._col2 (type: double), VALUE._col3 (type: double), VALUE._col4 (type: float), VALUE._col5 (type: double), VALUE._col6 (type: double), VALUE._col7 (type: double), VALUE._col8 (type: tinyint), VALUE._col9 (type: double), VALUE._col10 (type: float), VALUE._col11 (type: int), VALUE._col12 (type: decimal(13,2)), VALUE._col13 (type: double)
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20
-                Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                 File Output Operator
                   compressed: false
-                  Statistics: Num rows: 6144 Data size: 2592628 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 6104 Data size: 2575728 Basic stats: COMPLETE Column stats: COMPLETE
                   table:
                       input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                       output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_16.q.out b/ql/src/test/results/clientpositive/llap/vectorization_16.q.out
index 7e8cb81144fc..7de5092fc76d 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_16.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_16.q.out
@@ -94,7 +94,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3]
                         keys: _col0 (type: string), _col1 (type: double), _col2 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6
                         Statistics: Num rows: 5979 Data size: 825318 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_9.q.out b/ql/src/test/results/clientpositive/llap/vectorization_9.q.out
index 7e8cb81144fc..7de5092fc76d 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_9.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_9.q.out
@@ -94,7 +94,7 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3]
                         keys: _col0 (type: string), _col1 (type: double), _col2 (type: timestamp)
-                        minReductionHashAggr: 0.4
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6
                         Statistics: Num rows: 5979 Data size: 825318 Basic stats: COMPLETE Column stats: COMPLETE
diff --git a/ql/src/test/results/clientpositive/llap/vectorization_short_regress.q.out b/ql/src/test/results/clientpositive/llap/vectorization_short_regress.q.out
index da82903d7963..8f7d63935f59 100644
--- a/ql/src/test/results/clientpositive/llap/vectorization_short_regress.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorization_short_regress.q.out
@@ -2949,10 +2949,10 @@ STAGE PLANS:
                             vectorProcessingMode: HASH
                             projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
                         keys: _col0 (type: timestamp), _col1 (type: string)
-                        minReductionHashAggr: 0.5133463
+                        minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20, _col21, _col22
-                        Statistics: Num rows: 5980 Data size: 1579124 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 6144 Data size: 1622368 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: timestamp), _col1 (type: string)
                           null sort order: zz
@@ -2962,7 +2962,7 @@ STAGE PLANS:
                               className: VectorReduceSinkMultiKeyOperator
                               native: true
                               nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine tez IN [tez] IS true, No PTF TopN IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true
-                          Statistics: Num rows: 5980 Data size: 1579124 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 6144 Data size: 1622368 Basic stats: COMPLETE Column stats: COMPLETE
                           value expressions: _col2 (type: double), _col3 (type: double), _col4 (type: bigint), _col5 (type: bigint), _col6 (type: bigint), _col7 (type: bigint), _col8 (type: tinyint), _col9 (type: double), _col10 (type: double), _col11 (type: bigint), _col12 (type: double), _col13 (type: double), _col14 (type: bigint), _col15 (type: bigint), _col16 (type: bigint), _col17 (type: double), _col18 (type: bigint), _col19 (type: double), _col20 (type: double), _col21 (type: double), _col22 (type: bigint)
             Execution mode: vectorized, llap
             LLAP IO: all inputs
@@ -2997,7 +2997,7 @@ STAGE PLANS:
                 keys: KEY._col0 (type: timestamp), KEY._col1 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15, _col16, _col17, _col18, _col19, _col20, _col21, _col22
-                Statistics: Num rows: 5980 Data size: 1579124 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 5979 Data size: 1578826 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: _col0 (type: timestamp), _col1 (type: string), power(((_col2 - ((_col3 * _col3) / _col4)) / _col4), 0.5) (type: double), (UDFToDouble(_col5) / _col6) (type: double), _col7 (type: bigint), _col8 (type: tinyint), ((_col9 - ((_col10 * _col10) / _col11)) / if((_col11 = 1L), null, (_col11 - 1))) (type: double), ((_col12 - ((_col13 * _col13) / _col14)) / _col14) (type: double), (UDFToDouble(_col15) / _col16) (type: double), ((_col12 - ((_col13 * _col13) / _col14)) / if((_col14 = 1L), null, (_col14 - 1))) (type: double), (_col17 / _col18) (type: double), _col19 (type: double), ((_col9 - ((_col10 * _col10) / _col11)) / _col11) (type: double), power(((_col20 - ((_col21 * _col21) / _col22)) / _col22), 0.5) (type: double), _col15 (type: bigint)
                   outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14
@@ -3006,12 +3006,12 @@ STAGE PLANS:
                       native: true
                       projectedOutputColumnNums: [0, 1, 27, 29, 7, 8, 36, 40, 42, 49, 50, 19, 54, 59, 15]
                       selectExpressions: FuncPowerDoubleToDouble(col 26:double)(children: DoubleColDivideLongColumn(col 25:double, col 4:bigint)(children: DoubleColSubtractDoubleColumn(col 2:double, col 24:double)(children: DoubleColDivideLongColumn(col 23:double, col 4:bigint)(children: DoubleColMultiplyDoubleColumn(col 3:double, col 3:double) -> 23:double) -> 24:double) -> 25:double) -> 26:double) -> 27:double, DoubleColDivideLongColumn(col 28:double, col 6:bigint)(children: CastLongToDouble(col 5:bigint) -> 28:double) -> 29:double, DoubleColDivideLongColumn(col 32:double, col 35:bigint)(children: DoubleColSubtractDoubleColumn(col 9:double, col 31:double)(children: DoubleColDivideLongColumn(col 30:double, col 11:bigint)(children: DoubleColMultiplyDoubleColumn(col 10:double, col 10:double) -> 30:double) -> 31:double) -> 32:double, IfExprNullCondExpr(col 33:boolean, null, col 34:bigint)(children: LongColEqualLongScalar(col 11:bigint, val 1) -> 33:boolean, LongColSubtractLongScalar(col 11:bigint, val 1) -> 34:bigint) -> 35:bigint) -> 36:double, DoubleColDivideLongColumn(col 39:double, col 14:bigint)(children: DoubleColSubtractDoubleColumn(col 12:double, col 38:double)(children: DoubleColDivideLongColumn(col 37:double, col 14:bigint)(children: DoubleColMultiplyDoubleColumn(col 13:double, col 13:double) -> 37:double) -> 38:double) -> 39:double) -> 40:double, DoubleColDivideLongColumn(col 41:double, col 16:bigint)(children: CastLongToDouble(col 15:bigint) -> 41:double) -> 42:double, DoubleColDivideLongColumn(col 45:double, col 48:bigint)(children: DoubleColSubtractDoubleColumn(col 12:double, col 44:double)(children: DoubleColDivideLongColumn(col 43:double, col 14:bigint)(children: DoubleColMultiplyDoubleColumn(col 13:double, col 13:double) -> 43:double) -> 44:double) -> 45:double, IfExprNullCondExpr(col 46:boolean, null, col 47:bigint)(children: LongColEqualLongScalar(col 14:bigint, val 1) -> 46:boolean, LongColSubtractLongScalar(col 14:bigint, val 1) -> 47:bigint) -> 48:bigint) -> 49:double, DoubleColDivideLongColumn(col 17:double, col 18:bigint) -> 50:double, DoubleColDivideLongColumn(col 53:double, col 11:bigint)(children: DoubleColSubtractDoubleColumn(col 9:double, col 52:double)(children: DoubleColDivideLongColumn(col 51:double, col 11:bigint)(children: DoubleColMultiplyDoubleColumn(col 10:double, col 10:double) -> 51:double) -> 52:double) -> 53:double) -> 54:double, FuncPowerDoubleToDouble(col 58:double)(children: DoubleColDivideLongColumn(col 57:double, col 22:bigint)(children: DoubleColSubtractDoubleColumn(col 20:double, col 56:double)(children: DoubleColDivideLongColumn(col 55:double, col 22:bigint)(children: DoubleColMultiplyDoubleColumn(col 21:double, col 21:double) -> 55:double) -> 56:double) -> 57:double) -> 58:double) -> 59:double
-                  Statistics: Num rows: 5980 Data size: 1196404 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 5979 Data size: 1196170 Basic stats: COMPLETE Column stats: COMPLETE
                   Top N Key Operator
                     sort order: +++++++++++++++++++++++++++++++++++++++
                     keys: _col0 (type: timestamp), _col1 (type: string), _col2 (type: double), (_col2 * 10.175D) (type: double), (- _col2) (type: double), _col3 (type: double), (- _col2) (type: double), (-26.28D - _col2) (type: double), _col4 (type: bigint), (- _col4) (type: bigint), ((-26.28D - _col2) * (- _col2)) (type: double), _col5 (type: tinyint), (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4))) (type: double), (- (_col2 * 10.175D)) (type: double), _col6 (type: double), (_col6 + (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) (type: double), _col2 (type: double), (UDFToDouble((- _col4)) / _col2) (type: double), _col7 (type: double), (10.175D / _col3) (type: double), _col8 (type: double), _col9 (type: double), ((_col6 + (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) - (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) (type: double), (_col2 * 10.175D) (type: double), _col10 (type: double), (((_col6 + (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) - (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) * 10.175D) (type: double), (10.175D % (10.175D / _col3)) (type: double), (- _col5) (type: tinyint), _col11 (type: double), _col12 (type: double), (- ((-26.28D - _col2) * (- _col2))) (type: double), ((- _col2) % _col10) (type: double), (-26.28 / CAST( (- _col5) AS decimal(3,0))) (type: decimal(8,6)), _col13 (type: double), _col14 (type: bigint), ((_col6 + (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) / _col7) (type: double), _col4 (type: bigint), _col4 (type: bigint), ((_col6 + (((-26.28D - _col2) * (- _col2)) * UDFToDouble((- _col4)))) % -26.28D) (type: double)
                     null sort order: zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
-                    Statistics: Num rows: 5980 Data size: 1196404 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 5979 Data size: 1196170 Basic stats: COMPLETE Column stats: COMPLETE
                     top n: 50
                     Top N Key Vectorization:
                         className: VectorTopNKeyOperator
@@ -3025,7 +3025,7 @@ STAGE PLANS:
                           native: true
                           projectedOutputColumnNums: [0, 1, 27, 23, 24, 29, 25, 26, 7, 35, 31, 8, 30, 32, 36, 28, 27, 38, 40, 37, 42, 49, 41, 39, 50, 43, 45, 48, 19, 54, 44, 52, 145, 59, 15, 53, 7, 7, 55]
                           selectExpressions: DoubleColMultiplyDoubleScalar(col 27:double, val 10.175) -> 23:double, DoubleColUnaryMinus(col 27:double) -> 24:double, DoubleColUnaryMinus(col 27:double) -> 25:double, DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 26:double, LongColUnaryMinus(col 7:bigint) -> 35:bigint, DoubleColMultiplyDoubleColumn(col 28:double, col 30:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 28:double, DoubleColUnaryMinus(col 27:double) -> 30:double) -> 31:double, DoubleColMultiplyDoubleColumn(col 32:double, col 28:double)(children: DoubleColMultiplyDoubleColumn(col 28:double, col 30:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 28:double, DoubleColUnaryMinus(col 27:double) -> 30:double) -> 32:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 28:double) -> 30:double, DoubleColUnaryMinus(col 28:double)(children: DoubleColMultiplyDoubleScalar(col 27:double, val 10.175) -> 28:double) -> 32:double, DoubleColAddDoubleColumn(col 36:double, col 37:double)(children: DoubleColMultiplyDoubleColumn(col 38:double, col 28:double)(children: DoubleColMultiplyDoubleColumn(col 28:double, col 37:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 28:double, DoubleColUnaryMinus(col 27:double) -> 37:double) -> 38:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 28:double) -> 37:double) -> 28:double, DoubleColDivideDoubleColumn(col 37:double, col 27:double)(children: CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 37:double) -> 38:double, DoubleScalarDivideDoubleColumn(val 10.175, col 29:double) -> 37:double, DoubleColSubtractDoubleColumn(col 39:double, col 43:double)(children: DoubleColAddDoubleColumn(col 36:double, col 41:double)(children: DoubleColMultiplyDoubleColumn(col 43:double, col 39:double)(children: DoubleColMultiplyDoubleColumn(col 39:double, col 41:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 39:double, DoubleColUnaryMinus(col 27:double) -> 41:double) -> 43:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 39:double) -> 41:double) -> 39:double, DoubleColMultiplyDoubleColumn(col 44:double, col 41:double)(children: DoubleColMultiplyDoubleColumn(col 41:double, col 43:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 41:double, DoubleColUnaryMinus(col 27:double) -> 43:double) -> 44:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 41:double) -> 43:double) -> 41:double, DoubleColMultiplyDoubleScalar(col 27:double, val 10.175) -> 39:double, DoubleColMultiplyDoubleScalar(col 44:double, val 10.175)(children: DoubleColSubtractDoubleColumn(col 43:double, col 45:double)(children: DoubleColAddDoubleColumn(col 36:double, col 44:double)(children: DoubleColMultiplyDoubleColumn(col 45:double, col 43:double)(children: DoubleColMultiplyDoubleColumn(col 43:double, col 44:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 43:double, DoubleColUnaryMinus(col 27:double) -> 44:double) -> 45:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 43:double) -> 44:double) -> 43:double, DoubleColMultiplyDoubleColumn(col 51:double, col 44:double)(children: DoubleColMultiplyDoubleColumn(col 44:double, col 45:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 44:double, DoubleColUnaryMinus(col 27:double) -> 45:double) -> 51:double, CastLongToDouble(col 48:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 48:bigint) -> 44:double) -> 45:double) -> 44:double) -> 43:double, DoubleScalarModuloDoubleColumn(val 10.175, col 44:double)(children: DoubleScalarDivideDoubleColumn(val 10.175, col 29:double) -> 44:double) -> 45:double, LongColUnaryMinus(col 8:tinyint) -> 48:tinyint, DoubleColUnaryMinus(col 52:double)(children: DoubleColMultiplyDoubleColumn(col 44:double, col 51:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 44:double, DoubleColUnaryMinus(col 27:double) -> 51:double) -> 52:double) -> 44:double, DoubleColModuloDoubleColumn(col 51:double, col 50:double)(children: DoubleColUnaryMinus(col 27:double) -> 51:double) -> 52:double, DecimalScalarDivideDecimalColumn(val -26.28, col 127:decimal(3,0))(children: CastLongToDecimal(col 71:tinyint)(children: LongColUnaryMinus(col 8:tinyint) -> 71:tinyint) -> 127:decimal(3,0)) -> 145:decimal(8,6), DoubleColDivideDoubleColumn(col 51:double, col 40:double)(children: DoubleColAddDoubleColumn(col 36:double, col 53:double)(children: DoubleColMultiplyDoubleColumn(col 55:double, col 51:double)(children: DoubleColMultiplyDoubleColumn(col 51:double, col 53:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 51:double, DoubleColUnaryMinus(col 27:double) -> 53:double) -> 55:double, CastLongToDouble(col 71:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 71:bigint) -> 51:double) -> 53:double) -> 51:double) -> 53:double, DoubleColModuloDoubleScalar(col 51:double, val -26.28)(children: DoubleColAddDoubleColumn(col 36:double, col 55:double)(children: DoubleColMultiplyDoubleColumn(col 56:double, col 51:double)(children: DoubleColMultiplyDoubleColumn(col 51:double, col 55:double)(children: DoubleScalarSubtractDoubleColumn(val -26.28, col 27:double) -> 51:double, DoubleColUnaryMinus(col 27:double) -> 55:double) -> 56:double, CastLongToDouble(col 71:bigint)(children: LongColUnaryMinus(col 7:bigint) -> 71:bigint) -> 51:double) -> 55:double) -> 51:double) -> 55:double
-                      Statistics: Num rows: 5980 Data size: 2739514 Basic stats: COMPLETE Column stats: COMPLETE
+                      Statistics: Num rows: 5979 Data size: 2738988 Basic stats: COMPLETE Column stats: COMPLETE
                       Reduce Output Operator
                         key expressions: _col0 (type: timestamp), _col1 (type: string), _col2 (type: double), _col3 (type: double), _col4 (type: double), _col5 (type: double), _col6 (type: double), _col7 (type: double), _col8 (type: bigint), _col9 (type: bigint), _col10 (type: double), _col11 (type: tinyint), _col12 (type: double), _col13 (type: double), _col14 (type: double), _col15 (type: double), _col16 (type: double), _col17 (type: double), _col18 (type: double), _col19 (type: double), _col20 (type: double), _col21 (type: double), _col22 (type: double), _col23 (type: double), _col24 (type: double), _col25 (type: double), _col26 (type: double), _col27 (type: tinyint), _col28 (type: double), _col29 (type: double), _col30 (type: double), _col31 (type: double), _col32 (type: decimal(8,6)), _col33 (type: double), _col34 (type: bigint), _col35 (type: double), _col36 (type: bigint), _col37 (type: bigint), _col38 (type: double)
                         null sort order: zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
@@ -3034,7 +3034,7 @@ STAGE PLANS:
                             className: VectorReduceSinkObjectHashOperator
                             native: true
                             nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine tez IN [tez] IS true, No PTF TopN IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true
-                        Statistics: Num rows: 5980 Data size: 2739514 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 5979 Data size: 2738988 Basic stats: COMPLETE Column stats: COMPLETE
         Reducer 3 
             Execution mode: vectorized, llap
             Reduce Vectorization:
@@ -3051,7 +3051,7 @@ STAGE PLANS:
                     className: VectorSelectOperator
                     native: true
                     projectedOutputColumnNums: [0, 1, 2, 3, 4, 5, 4, 7, 8, 9, 10, 11, 12, 13, 14, 15, 2, 17, 18, 19, 20, 21, 22, 3, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 8, 8, 38]
-                Statistics: Num rows: 5980 Data size: 2739514 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 5979 Data size: 2738988 Basic stats: COMPLETE Column stats: COMPLETE
                 Limit
                   Number of rows: 50
                   Limit Vectorization:
diff --git a/ql/src/test/results/clientpositive/llap/vectorized_stats.q.out b/ql/src/test/results/clientpositive/llap/vectorized_stats.q.out
index af0c461861f3..0e1519cf20a9 100644
--- a/ql/src/test/results/clientpositive/llap/vectorized_stats.q.out
+++ b/ql/src/test/results/clientpositive/llap/vectorized_stats.q.out
@@ -1207,13 +1207,13 @@ STAGE PLANS:
                       minReductionHashAggr: 0.99
                       mode: hash
                       outputColumnNames: _col0
-                      Statistics: Num rows: 1 Data size: 40 Basic stats: COMPLETE Column stats: COMPLETE
+                      Statistics: Num rows: 6144 Data size: 183480 Basic stats: COMPLETE Column stats: COMPLETE
                       Reduce Output Operator
                         key expressions: _col0 (type: timestamp)
                         null sort order: z
                         sort order: +
                         Map-reduce partition columns: _col0 (type: timestamp)
-                        Statistics: Num rows: 1 Data size: 40 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 6144 Data size: 183480 Basic stats: COMPLETE Column stats: COMPLETE
             Execution mode: vectorized, llap
             LLAP IO: all inputs
         Reducer 2 
@@ -1223,10 +1223,10 @@ STAGE PLANS:
                 keys: KEY._col0 (type: timestamp)
                 mode: mergepartial
                 outputColumnNames: _col0
-                Statistics: Num rows: 1 Data size: 40 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 3072 Data size: 91760 Basic stats: COMPLETE Column stats: COMPLETE
                 File Output Operator
                   compressed: false
-                  Statistics: Num rows: 1 Data size: 40 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 3072 Data size: 91760 Basic stats: COMPLETE Column stats: COMPLETE
                   table:
                       input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                       output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat