cloud-fan commented on a change in pull request #26434: [SPARK-29544] [SQL] optimize skewed partition based on data size URL: https://github.com/apache/spark/pull/26434#discussion_r355987379
########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/SkewedShuffledRowRDD.scala ########## @@ -0,0 +1,91 @@ +/* + * 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.adaptive + +import org.apache.spark._ +import org.apache.spark.rdd.RDD +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.execution.metric.{SQLMetric, SQLShuffleReadMetricsReporter} + +/** + * The [[Partition]] used by [[SkewedShuffledRowRDD]]. A post-shuffle partition + * (identified by `postShufflePartitionIndex`) contains a range of pre-shuffle partitions + * (`preShufflePartitionIndex` from `startMapId` to `endMapId - 1`, inclusive). + */ +private final class SkewedShuffledRowRDDPartition( + val postShufflePartitionIndex: Int, + val preShufflePartitionIndex: Int, + val startMapId: Int, + val endMapId: Int) extends Partition{ + override val index: Int = postShufflePartitionIndex +} + +/** + * This is a specialized version of [[org.apache.spark.sql.execution.ShuffledRowRDD]]. This is used + * in Spark SQL adaptive execution to solve data skew issues. This RDD includes rearranged + * partitions from mappers. + * + * This RDD takes a [[ShuffleDependency]] (`dependency`), a partitionIndex + * and the range of startMapId to endMapId. + * + */ +class SkewedShuffledRowRDD( + var dependency: ShuffleDependency[Int, InternalRow, InternalRow], + partitionIndex: Int, + startMapId: Int, + endMapId: Int, + metrics: Map[String, SQLMetric]) + extends RDD[InternalRow](dependency.rdd.context, Nil) { + + override def getDependencies: Seq[Dependency[_]] = List(dependency) + override def getPartitions: Array[Partition] = { + Array.tabulate[Partition](1) { i => + new SkewedShuffledRowRDDPartition(i, partitionIndex, startMapId, endMapId) + } + } + + override def getPreferredLocations(partition: Partition): Seq[String] = { + val tracker = SparkEnv.get.mapOutputTracker.asInstanceOf[MapOutputTrackerMaster] + val skewedPartition = partition.asInstanceOf[SkewedShuffledRowRDDPartition] Review comment: all the partitions have the same preferred location, we don't even need to access the `partition` parameter. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
