Why does this Spark code raise a java.io.NotSerializableException

Question

I want to access the companion object method inside a conversion to RDD. Why the following does not work:

import org.apache.spark.rdd.RDD
import spark.implicits._
import org.apache.spark.sql.{Encoder, Encoders}

class Abc {
    def transform(x: RDD[Int]): RDD[Double] = { x.map(Abc.fn) }
}

object Abc {
  def fn(x: Int): Double = { x.toDouble }
}

implicit def abcEncoder: Encoder[Abc] = Encoders.kryo[Abc]

new Abc().transform(sc.parallelize(1 to 10)).collect

The above code snippet calls a java.io.NotSerializableException:

org.apache.spark.SparkException: Task not serializable
  at org.apache.spark.util.ClosureCleaner$.ensureSerializable(ClosureCleaner.scala:298)
  at org.apache.spark.util.ClosureCleaner$.org$apache$spark$util$ClosureCleaner$$clean(ClosureCleaner.scala:288)
  at org.apache.spark.util.ClosureCleaner$.clean(ClosureCleaner.scala:108)
  at org.apache.spark.SparkContext.clean(SparkContext.scala:2094)
  at org.apache.spark.rdd.RDD$$anonfun$map$1.apply(RDD.scala:370)
  at org.apache.spark.rdd.RDD$$anonfun$map$1.apply(RDD.scala:369)
  at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:151)
  at org.apache.spark.rdd.RDDOperationScope$.withScope(RDDOperationScope.scala:112)
  at org.apache.spark.rdd.RDD.withScope(RDD.scala:362)
  at org.apache.spark.rdd.RDD.map(RDD.scala:369)
  at Abc.transform(<console>:19)
  ... 47 elided
Caused by: java.io.NotSerializableException: Abc
Serialization stack:
        - object not serializable (class: Abc, value: Abc@4f598dfb)
        - field (class: Abc$$anonfun$transform$1, name: $outer, type: class Abc)
        - object (class Abc$$anonfun$transform$1, <function1>)
  at org.apache.spark.serializer.SerializationDebugger$.improveException(SerializationDebugger.scala:40)
  at org.apache.spark.serializer.JavaSerializationStream.writeObject(JavaSerializer.scala:46)
  at org.apache.spark.serializer.JavaSerializerInstance.serialize(JavaSerializer.scala:100)
  at org.apache.spark.util.ClosureCleaner$.ensureSerializable(ClosureCleaner.scala:295)
  ... 57 more

Even the definition Encoderfor the Abc class does not help here. But the more important question is: why is the serialization of an object of class Abc performed at all? My first thought was that the companion object is a single class object, so there may be an attempt to serialize it. But this is not like that, because when I call Abc.fn from another class:

class Xyz {
    def transform(x: RDD[Int]): RDD[Double] = { x.map(Abc.fn) }
}

implicit def xyzEncoder: Encoder[Xyz] = Encoders.kryo[Xyz]

new Xyz().transform(sc.parallelize(1 to 10)).collect

I get java.io.NotSerializableException: Xyz

Answer 1

, "" " Apache Spark:

Apache Spark,

:

• , ""

Answer 2

os - RDD, node . , RDD, node . .

ABC . ABC Serializable

class Abc with Serializable{
    def transform(x: RDD[Int]): RDD[Double] = { x.map(Abc.fn) }
}