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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
|
/*
* 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.rdd
import scala.reflect.ClassTag
import java.io.{ObjectOutputStream, IOException}
import org.apache.spark.{TaskContext, OneToOneDependency, SparkContext, Partition}
/**
* Class representing partitions of PartitionerAwareUnionRDD, which maintains the list of corresponding partitions
* of parent RDDs.
*/
private[spark]
class PartitionerAwareUnionRDDPartition(
@transient val rdds: Seq[RDD[_]],
val idx: Int
) extends Partition {
var parents = rdds.map(_.partitions(idx)).toArray
override val index = idx
override def hashCode(): Int = idx
@throws(classOf[IOException])
private def writeObject(oos: ObjectOutputStream) {
// Update the reference to parent partition at the time of task serialization
parents = rdds.map(_.partitions(index)).toArray
oos.defaultWriteObject()
}
}
/**
* Class representing an RDD that can take multiple RDDs partitioned by the same partitioner and
* unify them into a single RDD while preserving the partitioner. So m RDDs with p partitions each
* will be unified to a single RDD with p partitions and the same partitioner. The preferred
* location for each partition of the unified RDD will be the most common preferred location
* of the corresponding partitions of the parent RDDs. For example, location of partition 0
* of the unified RDD will be where most of partition 0 of the parent RDDs are located.
*/
private[spark]
class PartitionerAwareUnionRDD[T: ClassTag](
sc: SparkContext,
var rdds: Seq[RDD[T]]
) extends RDD[T](sc, rdds.map(x => new OneToOneDependency(x))) {
require(rdds.length > 0)
require(rdds.flatMap(_.partitioner).toSet.size == 1,
"Parent RDDs have different partitioners: " + rdds.flatMap(_.partitioner))
override val partitioner = rdds.head.partitioner
override def getPartitions: Array[Partition] = {
val numPartitions = partitioner.get.numPartitions
(0 until numPartitions).map(index => {
new PartitionerAwareUnionRDDPartition(rdds, index)
}).toArray
}
// Get the location where most of the partitions of parent RDDs are located
override def getPreferredLocations(s: Partition): Seq[String] = {
logDebug("Finding preferred location for " + this + ", partition " + s.index)
val parentPartitions = s.asInstanceOf[PartitionerAwareUnionRDDPartition].parents
val locations = rdds.zip(parentPartitions).flatMap {
case (rdd, part) => {
val parentLocations = currPrefLocs(rdd, part)
logDebug("Location of " + rdd + " partition " + part.index + " = " + parentLocations)
parentLocations
}
}
val location = if (locations.isEmpty) {
None
} else {
// Find the location that maximum number of parent partitions prefer
Some(locations.groupBy(x => x).maxBy(_._2.length)._1)
}
logDebug("Selected location for " + this + ", partition " + s.index + " = " + location)
location.toSeq
}
override def compute(s: Partition, context: TaskContext): Iterator[T] = {
val parentPartitions = s.asInstanceOf[PartitionerAwareUnionRDDPartition].parents
rdds.zip(parentPartitions).iterator.flatMap {
case (rdd, p) => rdd.iterator(p, context)
}
}
override def clearDependencies() {
super.clearDependencies()
rdds = null
}
// Get the *current* preferred locations from the DAGScheduler (as opposed to the static ones)
private def currPrefLocs(rdd: RDD[_], part: Partition): Seq[String] = {
rdd.context.getPreferredLocs(rdd, part.index).map(tl => tl.host)
}
}
|