1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
|
/*
* 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.exchange
import scala.collection.mutable
import scala.collection.mutable.ArrayBuffer
import org.apache.spark.broadcast
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.SQLContext
import org.apache.spark.sql.catalyst.InternalRow
import org.apache.spark.sql.catalyst.expressions.Attribute
import org.apache.spark.sql.catalyst.rules.Rule
import org.apache.spark.sql.execution.{LeafNode, SparkPlan, UnaryNode}
import org.apache.spark.sql.types.StructType
/**
* An interface for exchanges.
*/
abstract class Exchange extends UnaryNode {
override def output: Seq[Attribute] = child.output
}
/**
* A wrapper for reused exchange to have different output, because two exchanges which produce
* logically identical output will have distinct sets of output attribute ids, so we need to
* preserve the original ids because they're what downstream operators are expecting.
*/
case class ReusedExchange(override val output: Seq[Attribute], child: Exchange) extends LeafNode {
override def sameResult(plan: SparkPlan): Boolean = {
// Ignore this wrapper. `plan` could also be a ReusedExchange, so we reverse the order here.
plan.sameResult(child)
}
def doExecute(): RDD[InternalRow] = {
child.execute()
}
override protected[sql] def doExecuteBroadcast[T](): broadcast.Broadcast[T] = {
child.executeBroadcast()
}
// Do not repeat the same tree in explain.
override def treeChildren: Seq[SparkPlan] = Nil
}
/**
* Find out duplicated exchanges in the spark plan, then use the same exchange for all the
* references.
*/
private[sql] case class ReuseExchange(sqlContext: SQLContext) extends Rule[SparkPlan] {
def apply(plan: SparkPlan): SparkPlan = {
if (!sqlContext.conf.exchangeReuseEnabled) {
return plan
}
// Build a hash map using schema of exchanges to avoid O(N*N) sameResult calls.
val exchanges = mutable.HashMap[StructType, ArrayBuffer[Exchange]]()
plan.transformUp {
case exchange: Exchange =>
// the exchanges that have same results usually also have same schemas (same column names).
val sameSchema = exchanges.getOrElseUpdate(exchange.schema, ArrayBuffer[Exchange]())
val samePlan = sameSchema.find { e =>
exchange.sameResult(e)
}
if (samePlan.isDefined) {
// Keep the output of this exchange, the following plans require that to resolve
// attributes.
ReusedExchange(exchange.output, samePlan.get)
} else {
sameSchema += exchange
exchange
}
}
}
}
|