diff --git a/sql/core/src/main/scala/org/apache/spark/sql/KeyValueGroupedDataset.scala b/sql/core/src/main/scala/org/apache/spark/sql/KeyValueGroupedDataset.scala
index a6867a67eead..7e600286e955 100644
--- a/sql/core/src/main/scala/org/apache/spark/sql/KeyValueGroupedDataset.scala
+++ b/sql/core/src/main/scala/org/apache/spark/sql/KeyValueGroupedDataset.scala
@@ -25,6 +25,7 @@ import org.apache.spark.sql.catalyst.encoders.{encoderFor, ExpressionEncoder, Ou
 import org.apache.spark.sql.catalyst.expressions.{Alias, Attribute, CreateStruct}
 import org.apache.spark.sql.catalyst.plans.logical._
 import org.apache.spark.sql.execution.QueryExecution
+import org.apache.spark.sql.expressions.ReduceAggregator
 
 /**
  * :: Experimental ::
@@ -177,10 +178,10 @@ class KeyValueGroupedDataset[K, V] private[sql](
    * @since 1.6.0
    */
   def reduceGroups(f: (V, V) => V): Dataset[(K, V)] = {
-    val func = (key: K, it: Iterator[V]) => Iterator((key, it.reduce(f)))
+    val encoder = encoderFor[V]
+    val aggregator: TypedColumn[V, V] = new ReduceAggregator(f, encoder).toColumn
 
-    implicit val resultEncoder = ExpressionEncoder.tuple(kExprEnc, vExprEnc)
-    flatMapGroups(func)
+    agg(aggregator)
   }
 
   /**
diff --git a/sql/core/src/main/scala/org/apache/spark/sql/expressions/ReduceAggregator.scala b/sql/core/src/main/scala/org/apache/spark/sql/expressions/ReduceAggregator.scala
new file mode 100644
index 000000000000..0f013d49a2ca
--- /dev/null
+++ b/sql/core/src/main/scala/org/apache/spark/sql/expressions/ReduceAggregator.scala
@@ -0,0 +1,87 @@
+/*
+ * 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.spark.sql.expressions
+
+import org.apache.spark.annotation.Experimental
+import org.apache.spark.sql.Encoder
+import org.apache.spark.sql.catalyst.encoders.ExpressionEncoder
+
+/**
+ * :: Experimental ::
+ * A generic class for reduce aggregations, which accepts a reduce function that can be used to take
+ * all of the elements of a group and reduce them to a single value.
+ *
+ * @tparam T The input and output type for the reduce function.
+ * @param func The reduce aggregation function.
+ * @param encoder The encoder for the input and output type of the reduce function.
+ * @since 2.1.0
+ */
+@Experimental
+private[sql] class ReduceAggregator[T](func: (T, T) => T, encoder: ExpressionEncoder[T])
+    extends Aggregator[T, (Boolean, T), T] {
+
+  /**
+   * A zero value for this aggregation. It is represented as a Tuple2. The first element of the
+   * tuple is a false boolean value indicating the buffer is not initialized. The second element
+   * is initialized as a null value.
+   * @since 2.1.0
+   */
+  override def zero: (Boolean, T) = (false, null.asInstanceOf[T])
+
+  override def bufferEncoder: Encoder[(Boolean, T)] =
+    ExpressionEncoder.tuple(ExpressionEncoder[Boolean](), encoder)
+
+  override def outputEncoder: Encoder[T] = encoder
+
+  /**
+   * Combine two values to produce a new value. If the buffer `b` is not initialized, it simply
+   * takes the value of `a` and set the initialization flag to `true`.
+   * @since 2.1.0
+   */
+  override def reduce(b: (Boolean, T), a: T): (Boolean, T) = {
+    if (b._1) {
+      (true, func(b._2, a))
+    } else {
+      (true, a)
+    }
+  }
+
+  /**
+   * Merge two intermediate values. As it is possibly that the buffer is just the `zero` value
+   * coming from empty partition, it checks if the buffers are initialized, and only performs
+   * merging when they are initialized both.
+   * @since 2.1.0
+   */
+  override def merge(b1: (Boolean, T), b2: (Boolean, T)): (Boolean, T) = {
+    if (!b1._1) {
+      b2
+    } else if (!b2._1) {
+      b1
+    } else {
+      (true, func(b1._2, b2._2))
+    }
+  }
+
+  /**
+   * Transform the output of the reduction. Simply output the value in the buffer.
+   * @since 2.1.0
+   */
+  override def finish(reduction: (Boolean, T)): T = {
+    reduction._2
+  }
+}
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DatasetAggregatorSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DatasetAggregatorSuite.scala
index ddc4dcd2395b..535a6c7d21ed 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DatasetAggregatorSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DatasetAggregatorSuite.scala
@@ -20,7 +20,7 @@ package org.apache.spark.sql
 import scala.language.postfixOps
 
 import org.apache.spark.sql.catalyst.encoders.ExpressionEncoder
-import org.apache.spark.sql.expressions.Aggregator
+import org.apache.spark.sql.expressions.{Aggregator, ReduceAggregator}
 import org.apache.spark.sql.expressions.scalalang.typed
 import org.apache.spark.sql.functions._
 import org.apache.spark.sql.test.SharedSQLContext
@@ -314,4 +314,42 @@ class DatasetAggregatorSuite extends QueryTest with SharedSQLContext {
     val ds3 = sql("SELECT 'Some String' AS b, 1279869254 AS a").as[AggData]
     assert(ds3.select(NameAgg.toColumn).schema.head.nullable === true)
   }
+
+  test("ReduceAggregator: zero value") {
+    val encoder: ExpressionEncoder[Int] = ExpressionEncoder()
+    val func = (v1: Int, v2: Int) => v1 + v2
+    val aggregator: ReduceAggregator[Int] = new ReduceAggregator(func, encoder)
+    assert(aggregator.zero == (false, null))
+  }
+
+  test("ReduceAggregator: reduce, merge and finish") {
+    val encoder: ExpressionEncoder[Int] = ExpressionEncoder()
+    val func = (v1: Int, v2: Int) => v1 + v2
+    val aggregator: ReduceAggregator[Int] = new ReduceAggregator(func, encoder)
+
+    val firstReduce = aggregator.reduce(aggregator.zero, 1)
+    assert(firstReduce == (true, 1))
+
+    val secondReduce = aggregator.reduce(firstReduce, 2)
+    assert(secondReduce == (true, 3))
+
+    val thirdReduce = aggregator.reduce(secondReduce, 3)
+    assert(thirdReduce == (true, 6))
+
+    val mergeWithZero1 = aggregator.merge(aggregator.zero, firstReduce)
+    assert(mergeWithZero1 == (true, 1))
+
+    val mergeWithZero2 = aggregator.merge(secondReduce, aggregator.zero)
+    assert(mergeWithZero2 == (true, 3))
+
+    val mergeTwoReduced = aggregator.merge(firstReduce, secondReduce)
+    assert(mergeTwoReduced == (true, 4))
+
+    assert(aggregator.finish(firstReduce)== 1)
+    assert(aggregator.finish(secondReduce) == 3)
+    assert(aggregator.finish(thirdReduce) == 6)
+    assert(aggregator.finish(mergeWithZero1) == 1)
+    assert(aggregator.finish(mergeWithZero2) == 3)
+    assert(aggregator.finish(mergeTwoReduced) == 4)
+  }
 }