You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2021/10/17 00:48:26 UTC
[GitHub] [spark] viirya commented on a change in pull request #34298: [SPARK-34960][SQL] Aggregate push down for ORC
viirya commented on a change in pull request #34298:
URL: https://github.com/apache/spark/pull/34298#discussion_r730201130
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/orc/OrcColumnsStatistics.java
##########
@@ -0,0 +1,56 @@
+/*
+ * 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.execution.datasources.orc;
+
+import org.apache.orc.ColumnStatistics;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * Columns statistics interface wrapping ORC {@link ColumnStatistics}s.
+ *
+ * Because ORC {@link ColumnStatistics}s are stored as an flatten array in ORC file footer,
+ * this class is used to covert ORC {@link ColumnStatistics}s from array to nested tree structure,
+ * according to data types. This is used for aggregate push down in ORC.
+ */
+public class OrcColumnsStatistics {
Review comment:
I'm curious why this is in java?
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/orc/OrcColumnsStatistics.java
##########
@@ -0,0 +1,56 @@
+/*
+ * 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.execution.datasources.orc;
+
+import org.apache.orc.ColumnStatistics;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * Columns statistics interface wrapping ORC {@link ColumnStatistics}s.
+ *
+ * Because ORC {@link ColumnStatistics}s are stored as an flatten array in ORC file footer,
+ * this class is used to covert ORC {@link ColumnStatistics}s from array to nested tree structure,
+ * according to data types. This is used for aggregate push down in ORC.
+ */
+public class OrcColumnsStatistics {
Review comment:
I'm curious why this is in java? This doesn't look like to be an API open to others.
##########
File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/AggregatePushDownUtils.scala
##########
@@ -0,0 +1,141 @@
+/*
+ * 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.execution.datasources
+
+import org.apache.spark.sql.SparkSession
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.expressions.Expression
+import org.apache.spark.sql.connector.expressions.NamedReference
+import org.apache.spark.sql.connector.expressions.aggregate.{AggregateFunc, Aggregation, Count, CountStar, Max, Min}
+import org.apache.spark.sql.execution.RowToColumnConverter
+import org.apache.spark.sql.execution.vectorized.{OffHeapColumnVector, OnHeapColumnVector}
+import org.apache.spark.sql.types.{DataType, LongType, StructField, StructType}
+import org.apache.spark.sql.vectorized.{ColumnarBatch, ColumnVector}
+
+/**
+ * Utility class for aggregate push down to Parquet and ORC.
+ */
+object AggregatePushDownUtils {
+
+ /**
+ * Get the data schema for aggregate to be pushed down.
+ */
+ def getSchemaForPushedAggregation(
+ aggregation: Aggregation,
+ schema: StructType,
+ partitionNameSet: Set[String],
+ dataFilters: Seq[Expression],
+ isAllowedTypeForMinMaxAggregate: DataType => Boolean,
+ sparkSession: SparkSession): Option[StructType] = {
+
+ var finalSchema = new StructType()
+
+ def getStructFieldForCol(col: NamedReference): StructField = {
+ schema.apply(col.fieldNames.head)
+ }
+
+ def isPartitionCol(col: NamedReference) = {
+ partitionNameSet.contains(col.fieldNames.head)
+ }
+
+ def processMinOrMax(agg: AggregateFunc): Boolean = {
+ val (column, aggType) = agg match {
+ case max: Max => (max.column, "max")
+ case min: Min => (min.column, "min")
+ case _ =>
+ throw new IllegalArgumentException(s"Unexpected type of AggregateFunc ${agg.describe}")
+ }
+
+ if (isPartitionCol(column)) {
+ // don't push down partition column, footer doesn't have max/min for partition column
+ return false
+ }
+ val structField = getStructFieldForCol(column)
+
+ if (isAllowedTypeForMinMaxAggregate(structField.dataType)) {
+ finalSchema = finalSchema.add(structField.copy(s"$aggType(" + structField.name + ")"))
+ true
+ } else {
+ false
+ }
+ }
+
+ if (aggregation.groupByColumns.nonEmpty || dataFilters.nonEmpty) {
+ // Parquet/ORC footer has max/min/count for columns
+ // e.g. SELECT COUNT(col1) FROM t
+ // but footer doesn't have max/min/count for a column if max/min/count
+ // are combined with filter or group by
+ // e.g. SELECT COUNT(col1) FROM t WHERE col2 = 8
+ // SELECT COUNT(col1) FROM t GROUP BY col2
+ // Todo: 1. add support if groupby column is partition col
Review comment:
@huaxingao Don't you already add the support? No?
##########
File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/AggregatePushDownUtils.scala
##########
@@ -0,0 +1,141 @@
+/*
+ * 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.execution.datasources
+
+import org.apache.spark.sql.SparkSession
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.expressions.Expression
+import org.apache.spark.sql.connector.expressions.NamedReference
+import org.apache.spark.sql.connector.expressions.aggregate.{AggregateFunc, Aggregation, Count, CountStar, Max, Min}
+import org.apache.spark.sql.execution.RowToColumnConverter
+import org.apache.spark.sql.execution.vectorized.{OffHeapColumnVector, OnHeapColumnVector}
+import org.apache.spark.sql.types.{DataType, LongType, StructField, StructType}
+import org.apache.spark.sql.vectorized.{ColumnarBatch, ColumnVector}
+
+/**
+ * Utility class for aggregate push down to Parquet and ORC.
+ */
+object AggregatePushDownUtils {
+
+ /**
+ * Get the data schema for aggregate to be pushed down.
+ */
+ def getSchemaForPushedAggregation(
+ aggregation: Aggregation,
+ schema: StructType,
+ partitionNameSet: Set[String],
+ dataFilters: Seq[Expression],
+ isAllowedTypeForMinMaxAggregate: DataType => Boolean,
+ sparkSession: SparkSession): Option[StructType] = {
+
+ var finalSchema = new StructType()
+
+ def getStructFieldForCol(col: NamedReference): StructField = {
+ schema.apply(col.fieldNames.head)
+ }
+
+ def isPartitionCol(col: NamedReference) = {
+ partitionNameSet.contains(col.fieldNames.head)
+ }
+
+ def processMinOrMax(agg: AggregateFunc): Boolean = {
+ val (column, aggType) = agg match {
+ case max: Max => (max.column, "max")
+ case min: Min => (min.column, "min")
+ case _ =>
+ throw new IllegalArgumentException(s"Unexpected type of AggregateFunc ${agg.describe}")
+ }
+
+ if (isPartitionCol(column)) {
+ // don't push down partition column, footer doesn't have max/min for partition column
+ return false
+ }
+ val structField = getStructFieldForCol(column)
+
+ if (isAllowedTypeForMinMaxAggregate(structField.dataType)) {
+ finalSchema = finalSchema.add(structField.copy(s"$aggType(" + structField.name + ")"))
+ true
+ } else {
+ false
+ }
+ }
+
+ if (aggregation.groupByColumns.nonEmpty || dataFilters.nonEmpty) {
+ // Parquet/ORC footer has max/min/count for columns
+ // e.g. SELECT COUNT(col1) FROM t
+ // but footer doesn't have max/min/count for a column if max/min/count
+ // are combined with filter or group by
+ // e.g. SELECT COUNT(col1) FROM t WHERE col2 = 8
+ // SELECT COUNT(col1) FROM t GROUP BY col2
+ // Todo: 1. add support if groupby column is partition col
+ // (https://issues.apache.org/jira/browse/SPARK-36646)
+ // 2. add support if filter col is partition col
+ // (https://issues.apache.org/jira/browse/SPARK-36647)
+ return None
+ }
+
+ aggregation.groupByColumns.foreach { col =>
Review comment:
Hmm, doesn't `aggregation.groupByColumns` must be empty?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org