org.apache.sysds.runtime.instructions.spark.data

Class BlockPartitioner

java.lang.Object

org.apache.spark.Partitioner

org.apache.sysds.runtime.instructions.spark.data.BlockPartitioner

All Implemented Interfaces:

Serializable

public class BlockPartitioner
extends org.apache.spark.Partitioner

Default partitioner used for all binary block rdd operations in order to enable sufficient local aggregation independent of the aggregation direction (row/col-wise). Especially, on large squared matrices (as common for factorization or graph algorithms), this is crucial for performance.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
BlockPartitioner(DataCharacteristics mc, int numParts)

Method Summary

Modifier and Type	Method and Description
boolean	equals(Object obj)
int	getPartition(Object arg0)
int	hashCode()
int	numPartitions()

Methods inherited from class org.apache.spark.Partitioner

defaultPartitioner

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BlockPartitioner

public BlockPartitioner(DataCharacteristics mc,
                        int numParts)

Method Detail

getPartition

public int getPartition(Object arg0)

Specified by:

getPartition in class org.apache.spark.Partitioner

numPartitions

public int numPartitions()

Specified by:

numPartitions in class org.apache.spark.Partitioner

hashCode

public int hashCode()

Overrides:

hashCode in class Object

equals

public boolean equals(Object obj)

Overrides:

equals in class Object