How can I take a rdd array of spark, and split it into two rdds randomly so each rdd will include some part of data (lets say 97% and 3%).
I thought to shuffle the list and then shuffledList.take((0.97*rddList.count).toInt)
But how can I Shuffle the rdd?
Or is there a better way to split the list?
I've found a simple and fast way to split the array:
val Array(f1,f2) = data.randomSplit(Array(0.97, 0.03))
It will split the data using the provided weights.
You should use randomSplit method:
def randomSplit(weights: Array[Double], seed: Long = Utils.random.nextLong): Array[RDD[T]]
// Randomly splits this RDD with the provided weights.
// weights for splits, will be normalized if they don't sum to 1
// returns split RDDs in an array
Here is its implementation in spark 1.0:
def randomSplit(weights: Array[Double], seed: Long = Utils.random.nextLong): Array[RDD[T]] = {
val sum = weights.sum
val normalizedCumWeights = weights.map(_ / sum).scanLeft(0.0d)(_ + _)
normalizedCumWeights.sliding(2).map { x =>
new PartitionwiseSampledRDD[T, T](this, new BernoulliSampler[T](x(0), x(1)),seed)
}.toArray
}
Related
I am wondering how to filter an RDD that has one of the top N values. Usually I would sort the RDD and take the top N items as an array in the driver to find the Nth value that can be broadcasted to filter the rdd like so:
val topNvalues = sc.broadcast(rdd.map(_.fieldToThreshold).distict.sorted.take(N))
val threshold = topNvalues.last
val rddWithTopNValues = rdd.filter(_.fieldToThreshold >= threshold)
but in this case my N is too large, so how can I do this purely with RDDs like so?:
def getExpensiveItems(itemPrices: RDD[(Int, Float)], count: Int): RDD[(Int, Float)] = {
val sortedPrices = itemPrices.sortBy(-_._2).map(_._1).distinct
// How to do this without collecting results to driver??
val highPrices = itemPrices.getTopNValuesWithoutCollect(count)
itemPrices.join(highPrices.keyBy(x => x)).map(_._2._1)
}
Use zipWithIndex on the sorted rdd and then filter by the index up to n items. To illustrate the case consider this rrd sorted in descending order,
val rdd = sc.parallelize((1 to 10).map( _ => math.random)).sortBy(-_)
Then
rdd.zipWithIndex.filter(_._2 < 4)
delivers the first top four items without collecting the rdd to the driver.
I have the below code which generates RDD from a text file:
val data = sparkContext.textfile(path)
val k = 3
How can I divide data into k unique RDD?
You can use RDD.randomSplitwhich will divide existing RDD based on weights passed in the parameters and return Array of RDDs.
The internal working will be like below...
/**
* Randomly splits this RDD with the provided weights.
*
* #param weights weights for splits, will be normalized if they don't sum to 1
* #param seed random seed
*
* #return split RDDs in an array
*/
def randomSplit(weights: Array[Double], seed: Long = Utils.random.nextLong): Array[RDD[T]] = {
require(weights.forall(_ >= 0),
s"Weights must be nonnegative, but got ${weights.mkString("[", ",", "]")}")
require(weights.sum > 0,
s"Sum of weights must be positive, but got ${weights.mkString("[", ",", "]")}")
withScope {
val sum = weights.sum
val normalizedCumWeights = weights.map(_ / sum).scanLeft(0.0d)(_ + _)
normalizedCumWeights.sliding(2).map { x =>
randomSampleWithRange(x(0), x(1), seed)
}.toArray
}
NOTE : weights weights for splits, will be normalized if they don't sum to 1
Based on the above behavior I created a sample snippet like below which was working :
def getDoubleWeights(numparts:Int) : Array[Double] = {
Array.fill[Double](numparts)(1.0d)
}
caller would be like....
val rddWithNumParts : Array[RDD] = yourRDD.randomSplit(getDoubleWeights(yourRDD.partitions.length))
This will uniformly divide in to number of RDD
NOTE : Same is applicable for below DataFrame.randomSplit as well
You can also convert that in to Dataframe by giving schema to RDD and use like below example.. sqlContext.createDataFrame(rddOfRow, Schema)
later you can call this method.
DataFrame[] randomSplit(double[] weights) Randomly splits this
DataFrame with the provided weights.
other thought I had is dividing based on number of Partitions...
i.e RDD.mapPartitionWithIndex(....)
for each partition you have an Iterator (can be converted in to RDD). you can have some thing like number of partitions = number of RDDs
I'm looking for a way to split an RDD into two or more RDDs. The closest I've seen is Scala Spark: Split collection into several RDD? which is still a single RDD.
If you're familiar with SAS, something like this:
data work.split1, work.split2;
set work.preSplit;
if (condition1)
output work.split1
else if (condition2)
output work.split2
run;
which resulted in two distinct data sets. It would have to be immediately persisted to get the results I intend...
It is not possible to yield multiple RDDs from a single transformation*. If you want to split a RDD you have to apply a filter for each split condition. For example:
def even(x): return x % 2 == 0
def odd(x): return not even(x)
rdd = sc.parallelize(range(20))
rdd_odd, rdd_even = (rdd.filter(f) for f in (odd, even))
If you have only a binary condition and computation is expensive you may prefer something like this:
kv_rdd = rdd.map(lambda x: (x, odd(x)))
kv_rdd.cache()
rdd_odd = kv_rdd.filter(lambda kv: kv[1]).keys()
rdd_even = kv_rdd.filter(lambda kv: not kv[1]).keys()
It means only a single predicate computation but requires additional pass over all data.
It is important to note that as long as an input RDD is properly cached and there no additional assumptions regarding data distribution there is no significant difference when it comes to time complexity between repeated filter and for-loop with nested if-else.
With N elements and M conditions number of operations you have to perform is clearly proportional to N times M. In case of for-loop it should be closer to (N + MN) / 2 and repeated filter is exactly NM but at the end of the day it is nothing else than O(NM). You can see my discussion** with Jason Lenderman to read about some pros-and-cons.
At the very high level you should consider two things:
Spark transformations are lazy, until you execute an action your RDD is not materialized
Why does it matter? Going back to my example:
rdd_odd, rdd_even = (rdd.filter(f) for f in (odd, even))
If later I decide that I need only rdd_odd then there is no reason to materialize rdd_even.
If you take a look at your SAS example to compute work.split2 you need to materialize both input data and work.split1.
RDDs provide a declarative API. When you use filter or map it is completely up to Spark engine how this operation is performed. As long as the functions passed to transformations are side effects free it creates multiple possibilities to optimize a whole pipeline.
At the end of the day this case is not special enough to justify its own transformation.
This map with filter pattern is actually used in a core Spark. See my answer to How does Sparks RDD.randomSplit actually split the RDD and a relevant part of the randomSplit method.
If the only goal is to achieve a split on input it is possible to use partitionBy clause for DataFrameWriter which text output format:
def makePairs(row: T): (String, String) = ???
data
.map(makePairs).toDF("key", "value")
.write.partitionBy($"key").format("text").save(...)
* There are only 3 basic types of transformations in Spark:
RDD[T] => RDD[T]
RDD[T] => RDD[U]
(RDD[T], RDD[U]) => RDD[W]
where T, U, W can be either atomic types or products / tuples (K, V). Any other operation has to be expressed using some combination of the above. You can check the original RDD paper for more details.
** https://chat.stackoverflow.com/rooms/91928/discussion-between-zero323-and-jason-lenderman
*** See also Scala Spark: Split collection into several RDD?
As other posters mentioned above, there is no single, native RDD transform that splits RDDs, but here are some "multiplex" operations that can efficiently emulate a wide variety of "splitting" on RDDs, without reading multiple times:
http://silex.freevariable.com/latest/api/#com.redhat.et.silex.rdd.multiplex.MuxRDDFunctions
Some methods specific to random splitting:
http://silex.freevariable.com/latest/api/#com.redhat.et.silex.sample.split.SplitSampleRDDFunctions
Methods are available from open source silex project:
https://github.com/willb/silex
A blog post explaining how they work:
http://erikerlandson.github.io/blog/2016/02/08/efficient-multiplexing-for-spark-rdds/
def muxPartitions[U :ClassTag](n: Int, f: (Int, Iterator[T]) => Seq[U],
persist: StorageLevel): Seq[RDD[U]] = {
val mux = self.mapPartitionsWithIndex { case (id, itr) =>
Iterator.single(f(id, itr))
}.persist(persist)
Vector.tabulate(n) { j => mux.mapPartitions { itr => Iterator.single(itr.next()(j)) } }
}
def flatMuxPartitions[U :ClassTag](n: Int, f: (Int, Iterator[T]) => Seq[TraversableOnce[U]],
persist: StorageLevel): Seq[RDD[U]] = {
val mux = self.mapPartitionsWithIndex { case (id, itr) =>
Iterator.single(f(id, itr))
}.persist(persist)
Vector.tabulate(n) { j => mux.mapPartitions { itr => itr.next()(j).toIterator } }
}
As mentioned elsewhere, these methods do involve a trade-off of memory for speed, because they operate by computing entire partition results "eagerly" instead of "lazily." Therefore, it is possible for these methods to run into memory problems on large partitions, where more traditional lazy transforms will not.
One way is to use a custom partitioner to partition the data depending upon your filter condition. This can be achieved by extending Partitioner and implementing something similar to the RangePartitioner.
A map partitions can then be used to construct multiple RDDs from the partitioned RDD without reading all the data.
val filtered = partitioned.mapPartitions { iter => {
new Iterator[Int](){
override def hasNext: Boolean = {
if(rangeOfPartitionsToKeep.contains(TaskContext.get().partitionId)) {
false
} else {
iter.hasNext
}
}
override def next():Int = iter.next()
}
Just be aware that the number of partitions in the filtered RDDs will be the same as the number in the partitioned RDD so a coalesce should be used to reduce this down and remove the empty partitions.
If you split an RDD using the randomSplit API call, you get back an array of RDDs.
If you want 5 RDDs returned, pass in 5 weight values.
e.g.
val sourceRDD = val sourceRDD = sc.parallelize(1 to 100, 4)
val seedValue = 5
val splitRDD = sourceRDD.randomSplit(Array(1.0,1.0,1.0,1.0,1.0), seedValue)
splitRDD(1).collect()
res7: Array[Int] = Array(1, 6, 11, 12, 20, 29, 40, 62, 64, 75, 77, 83, 94, 96, 100)
scala spark : How to avoid RDD shuffling in join after Distributed Matrix operation
created a dense matrix as a input to calculate cosine distance between columns
val rowMarixIn = sc.textFile("input.csv").map{ line =>
val values = line.split(" ").map(_.toDouble)
Vectors.dense(values)
}
Extracted set of entries from co-ordinated matrix after the cosine calculations
val coMarix = new RowMatrix(rowMarixIn)
val similerRows = coMatrix.columnSimilarities()
//extract entires over a specific Threshold
val rowIndices = similerRows.entries.map {case MatrixEntry(row: Long, col: Long, sim: Double) =>
if (sim > someTreshold )){
col,sim
}`
We have a another RDD with rdd2(key,Val2)
just want to join the two rdd's, rowIndices(key,Val) , rdd2(key,Val2)
val joinedRDD = rowIndices.join(rdd2)
this will result in a shuffle ,
What are best practices to follow in order to avoid shuffle or any suggestion on a better approach is much appreciated
I'm looking for a way to split an RDD into two or more RDDs. The closest I've seen is Scala Spark: Split collection into several RDD? which is still a single RDD.
If you're familiar with SAS, something like this:
data work.split1, work.split2;
set work.preSplit;
if (condition1)
output work.split1
else if (condition2)
output work.split2
run;
which resulted in two distinct data sets. It would have to be immediately persisted to get the results I intend...
It is not possible to yield multiple RDDs from a single transformation*. If you want to split a RDD you have to apply a filter for each split condition. For example:
def even(x): return x % 2 == 0
def odd(x): return not even(x)
rdd = sc.parallelize(range(20))
rdd_odd, rdd_even = (rdd.filter(f) for f in (odd, even))
If you have only a binary condition and computation is expensive you may prefer something like this:
kv_rdd = rdd.map(lambda x: (x, odd(x)))
kv_rdd.cache()
rdd_odd = kv_rdd.filter(lambda kv: kv[1]).keys()
rdd_even = kv_rdd.filter(lambda kv: not kv[1]).keys()
It means only a single predicate computation but requires additional pass over all data.
It is important to note that as long as an input RDD is properly cached and there no additional assumptions regarding data distribution there is no significant difference when it comes to time complexity between repeated filter and for-loop with nested if-else.
With N elements and M conditions number of operations you have to perform is clearly proportional to N times M. In case of for-loop it should be closer to (N + MN) / 2 and repeated filter is exactly NM but at the end of the day it is nothing else than O(NM). You can see my discussion** with Jason Lenderman to read about some pros-and-cons.
At the very high level you should consider two things:
Spark transformations are lazy, until you execute an action your RDD is not materialized
Why does it matter? Going back to my example:
rdd_odd, rdd_even = (rdd.filter(f) for f in (odd, even))
If later I decide that I need only rdd_odd then there is no reason to materialize rdd_even.
If you take a look at your SAS example to compute work.split2 you need to materialize both input data and work.split1.
RDDs provide a declarative API. When you use filter or map it is completely up to Spark engine how this operation is performed. As long as the functions passed to transformations are side effects free it creates multiple possibilities to optimize a whole pipeline.
At the end of the day this case is not special enough to justify its own transformation.
This map with filter pattern is actually used in a core Spark. See my answer to How does Sparks RDD.randomSplit actually split the RDD and a relevant part of the randomSplit method.
If the only goal is to achieve a split on input it is possible to use partitionBy clause for DataFrameWriter which text output format:
def makePairs(row: T): (String, String) = ???
data
.map(makePairs).toDF("key", "value")
.write.partitionBy($"key").format("text").save(...)
* There are only 3 basic types of transformations in Spark:
RDD[T] => RDD[T]
RDD[T] => RDD[U]
(RDD[T], RDD[U]) => RDD[W]
where T, U, W can be either atomic types or products / tuples (K, V). Any other operation has to be expressed using some combination of the above. You can check the original RDD paper for more details.
** https://chat.stackoverflow.com/rooms/91928/discussion-between-zero323-and-jason-lenderman
*** See also Scala Spark: Split collection into several RDD?
As other posters mentioned above, there is no single, native RDD transform that splits RDDs, but here are some "multiplex" operations that can efficiently emulate a wide variety of "splitting" on RDDs, without reading multiple times:
http://silex.freevariable.com/latest/api/#com.redhat.et.silex.rdd.multiplex.MuxRDDFunctions
Some methods specific to random splitting:
http://silex.freevariable.com/latest/api/#com.redhat.et.silex.sample.split.SplitSampleRDDFunctions
Methods are available from open source silex project:
https://github.com/willb/silex
A blog post explaining how they work:
http://erikerlandson.github.io/blog/2016/02/08/efficient-multiplexing-for-spark-rdds/
def muxPartitions[U :ClassTag](n: Int, f: (Int, Iterator[T]) => Seq[U],
persist: StorageLevel): Seq[RDD[U]] = {
val mux = self.mapPartitionsWithIndex { case (id, itr) =>
Iterator.single(f(id, itr))
}.persist(persist)
Vector.tabulate(n) { j => mux.mapPartitions { itr => Iterator.single(itr.next()(j)) } }
}
def flatMuxPartitions[U :ClassTag](n: Int, f: (Int, Iterator[T]) => Seq[TraversableOnce[U]],
persist: StorageLevel): Seq[RDD[U]] = {
val mux = self.mapPartitionsWithIndex { case (id, itr) =>
Iterator.single(f(id, itr))
}.persist(persist)
Vector.tabulate(n) { j => mux.mapPartitions { itr => itr.next()(j).toIterator } }
}
As mentioned elsewhere, these methods do involve a trade-off of memory for speed, because they operate by computing entire partition results "eagerly" instead of "lazily." Therefore, it is possible for these methods to run into memory problems on large partitions, where more traditional lazy transforms will not.
One way is to use a custom partitioner to partition the data depending upon your filter condition. This can be achieved by extending Partitioner and implementing something similar to the RangePartitioner.
A map partitions can then be used to construct multiple RDDs from the partitioned RDD without reading all the data.
val filtered = partitioned.mapPartitions { iter => {
new Iterator[Int](){
override def hasNext: Boolean = {
if(rangeOfPartitionsToKeep.contains(TaskContext.get().partitionId)) {
false
} else {
iter.hasNext
}
}
override def next():Int = iter.next()
}
Just be aware that the number of partitions in the filtered RDDs will be the same as the number in the partitioned RDD so a coalesce should be used to reduce this down and remove the empty partitions.
If you split an RDD using the randomSplit API call, you get back an array of RDDs.
If you want 5 RDDs returned, pass in 5 weight values.
e.g.
val sourceRDD = val sourceRDD = sc.parallelize(1 to 100, 4)
val seedValue = 5
val splitRDD = sourceRDD.randomSplit(Array(1.0,1.0,1.0,1.0,1.0), seedValue)
splitRDD(1).collect()
res7: Array[Int] = Array(1, 6, 11, 12, 20, 29, 40, 62, 64, 75, 77, 83, 94, 96, 100)