Github user marmbrus commented on a diff in the pull request:
https://github.com/apache/spark/pull/7379#discussion_r34940933
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/joins/BroadcastRangeJoin.scala
---
@@ -0,0 +1,411 @@
+/*
+ * 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.joins
+
+import org.apache.spark.sql.catalyst.util.TypeUtils
+import org.apache.spark.util.ThreadUtils
+
+import scala.annotation.tailrec
+import scala.collection.mutable
+import scala.concurrent._
+import scala.concurrent.duration._
+
+import org.apache.spark.annotation.DeveloperApi
+import org.apache.spark.rdd.RDD
+import org.apache.spark.sql.Row
+import org.apache.spark.sql.catalyst.InternalRow
+import org.apache.spark.sql.catalyst.expressions._
+import org.apache.spark.sql.execution.{BinaryNode, SparkPlan}
+
+/**
+ * Performs an inner range join on two tables. A range join typically has
the following form:
+ *
+ * SELECT A.*
+ * ,B.*
+ * FROM tableA A
+ * JOIN tableB B
+ * ON A.start <= B.end
+ * AND A.end > B.start
+ *
+ * The implementation builds a range index from the smaller build side,
broadcasts this index
+ * to all executors. The streaming side is then matched against the index.
This reduces the number
+ * of comparisons made by log(n) (n is the number of records in the build
table) over the
+ * typical solution (Nested Loop Join).
+ *
+ * TODO NaN values
+ * TODO NULL values
+ * TODO Outer joins? StreamSide is quite easy/BuildSide requires
bookkeeping and
+ * TODO This join will maintain sort order. The build side rows will also
be added in a lower
+ * bound sorted fashion.
+ */
+@DeveloperApi
+case class BroadcastRangeJoin(
+ leftKeys: Seq[Expression],
+ rightKeys: Seq[Expression],
+ equality: Seq[Boolean],
+ buildSide: BuildSide,
+ left: SparkPlan,
+ right: SparkPlan)
+ extends BinaryNode {
+
+ private[this] lazy val (buildPlan, streamedPlan) = buildSide match {
+ case BuildLeft => (left, right)
+ case BuildRight => (right, left)
+ }
+
+ private[this] lazy val (buildKeys, streamedKeys) = buildSide match {
+ case BuildLeft => (leftKeys, rightKeys)
+ case BuildRight => (rightKeys, leftKeys)
+ }
+
+ override def output: Seq[Attribute] = left.output ++ right.output
+
+ @transient
+ private[this] lazy val buildSideKeyGenerator: Projection =
+ newProjection(buildKeys, buildPlan.output)
+
+ @transient
+ private[this] lazy val streamSideKeyGenerator: () => MutableProjection =
+ newMutableProjection(streamedKeys, streamedPlan.output)
+
+ private[this] val timeout: Duration = {
--- End diff --
maybe we should just put this logic in `SQLConf` so we can use it in both
places.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
