org.apache.beam.runners.spark.io

Class SourceRDD.Unbounded<T,CheckpointMarkT extends UnboundedSource.CheckpointMark>

java.lang.Object

org.apache.spark.rdd.RDD<scala.Tuple2<Source<T>,CheckpointMarkT>>

org.apache.beam.runners.spark.io.SourceRDD.Unbounded<T,CheckpointMarkT>

All Implemented Interfaces:

java.io.Serializable, org.apache.spark.internal.Logging

Enclosing class:

SourceRDD

public static class SourceRDD.Unbounded<T,CheckpointMarkT extends UnboundedSource.CheckpointMark>
extends org.apache.spark.rdd.RDD<scala.Tuple2<Source<T>,CheckpointMarkT>>

A SourceRDD.Unbounded is the implementation of a micro-batch in a SourceDStream.

This RDD is made of P partitions, each containing a single pair-element of the partitioned MicrobatchSource and an optional starting UnboundedSource.CheckpointMark.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
Unbounded(org.apache.spark.SparkContext sc, org.apache.beam.runners.core.construction.SerializablePipelineOptions options, MicrobatchSource<T,CheckpointMarkT> microbatchSource, int initialNumPartitions)

Method Summary

Modifier and Type	Method and Description
scala.collection.Iterator<scala.Tuple2<Source<T>,CheckpointMarkT>>	compute(org.apache.spark.Partition split, org.apache.spark.TaskContext context)
org.apache.spark.Partition[]	getPartitions()
scala.Option<org.apache.spark.Partitioner>	partitioner()

Methods inherited from class org.apache.spark.rdd.RDD

$plus$plus, aggregate, barrier, cache, cartesian, checkpoint, checkpointData_$eq, checkpointData, cleanShuffleDependencies, cleanShuffleDependencies$default$1, clearDependencies, coalesce, coalesce$default$2, coalesce$default$3, coalesce$default$4, collect, collect, collectPartitions, computeOrReadCheckpoint, conf, context, count, countApprox, countApprox$default$2, countApproxDistinct, countApproxDistinct, countApproxDistinct$default$1, countByValue, countByValue$default$1, countByValueApprox, countByValueApprox$default$2, countByValueApprox$default$3, creationSite, dependencies, distinct, distinct, distinct$default$2, doCheckpoint, doubleRDDToDoubleRDDFunctions, elementClassTag, filter, first, firstParent, flatMap, fold, foreach, foreachPartition, getCheckpointFile, getCreationSite, getDependencies, getNarrowAncestors, getNumPartitions, getOrCompute, getOutputDeterministicLevel, getPreferredLocations, getResourceProfile, getStorageLevel, glom, groupBy, groupBy, groupBy, groupBy$default$4, id, initializeForcefully, initializeLogIfNecessary, initializeLogIfNecessary, initializeLogIfNecessary$default$2, intersection, intersection, intersection, intersection$default$3, isBarrier_, isBarrier, isCheckpointed, isCheckpointedAndMaterialized, isEmpty, isLocallyCheckpointed, isReliablyCheckpointed, isTraceEnabled, iterator, keyBy, localCheckpoint, log, logDebug, logDebug, logError, logError, logInfo, logInfo, logName, logTrace, logTrace, logWarning, logWarning, map, mapPartitions, mapPartitions$default$2, mapPartitionsInternal, mapPartitionsInternal$default$2, mapPartitionsWithIndex, mapPartitionsWithIndex, mapPartitionsWithIndex$default$2, mapPartitionsWithIndexInternal, mapPartitionsWithIndexInternal$default$2, mapPartitionsWithIndexInternal$default$3, markCheckpointed, max, min, name_$eq, name, numericRDDToDoubleRDDFunctions, org$apache$spark$internal$Logging$$log__$eq, org$apache$spark$internal$Logging$$log_, outputDeterministicLevel, parent, partitions, persist, persist, pipe, pipe, pipe, pipe$default$2, pipe$default$3, pipe$default$4, pipe$default$5, pipe$default$6, pipe$default$7, preferredLocations, randomSampleWithRange, randomSplit, randomSplit$default$2, rddToAsyncRDDActions, rddToOrderedRDDFunctions, rddToPairRDDFunctions, rddToPairRDDFunctions$default$4, rddToSequenceFileRDDFunctions, reduce, repartition, repartition$default$2, retag, retag, sample, sample$default$3, saveAsObjectFile, saveAsTextFile, saveAsTextFile, scope, setName, sortBy, sortBy$default$2, sortBy$default$3, sparkContext, subtract, subtract, subtract, subtract$default$3, take, takeOrdered, takeSample, takeSample$default$3, toDebugString, toJavaRDD, toLocalIterator, top, toString, treeAggregate, treeAggregate$default$4, treeReduce, treeReduce$default$2, union, unpersist, unpersist$default$1, withResources, withScope, zip, zipPartitions, zipPartitions, zipPartitions, zipPartitions, zipPartitions, zipPartitions, zipWithIndex, zipWithUniqueId

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.apache.spark.internal.Logging

$init$, initLock, uninitialize

Constructor Detail

Unbounded

public Unbounded(org.apache.spark.SparkContext sc,
                 org.apache.beam.runners.core.construction.SerializablePipelineOptions options,
                 MicrobatchSource<T,CheckpointMarkT> microbatchSource,
                 int initialNumPartitions)

Method Detail

getPartitions

public org.apache.spark.Partition[] getPartitions()

Specified by:

getPartitions in class org.apache.spark.rdd.RDD<scala.Tuple2<Source<T>,CheckpointMarkT extends UnboundedSource.CheckpointMark>>

partitioner

public scala.Option<org.apache.spark.Partitioner> partitioner()

Overrides:

partitioner in class org.apache.spark.rdd.RDD<scala.Tuple2<Source<T>,CheckpointMarkT extends UnboundedSource.CheckpointMark>>

compute

public scala.collection.Iterator<scala.Tuple2<Source<T>,CheckpointMarkT>> compute(org.apache.spark.Partition split,
                                                                                  org.apache.spark.TaskContext context)

Specified by:

compute in class org.apache.spark.rdd.RDD<scala.Tuple2<Source<T>,CheckpointMarkT extends UnboundedSource.CheckpointMark>>