org.apache.spark.storage

Class StorageUtils

Object

org.apache.spark.storage.StorageUtils

public class StorageUtils
extends Object

Helper methods for storage-related objects.

Constructor Summary

Constructors

Constructor and Description
StorageUtils()

Method Summary

Methods

Modifier and Type	Method and Description
static void	dispose(java.nio.ByteBuffer buffer) Attempt to clean up a ByteBuffer if it is memory-mapped.
static scala.collection.Map<BlockId,scala.collection.Seq<String>>	getRddBlockLocations(int rddId, scala.collection.Seq<StorageStatus> statuses) Return a mapping from block ID to its locations for each block that belongs to the given RDD.
static void	updateRddInfo(scala.collection.Seq<RDDInfo> rddInfos, scala.collection.Seq<StorageStatus> statuses) Update the given list of RDDInfo with the given list of storage statuses.

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

StorageUtils

public StorageUtils()

Method Detail

dispose

public static void dispose(java.nio.ByteBuffer buffer)

Attempt to clean up a ByteBuffer if it is memory-mapped. This uses an *unsafe* Sun API that might cause errors if one attempts to read from the unmapped buffer, but it's better than waiting for the GC to find it because that could lead to huge numbers of open files. There's unfortunately no standard API to do this.

Parameters:

buffer - (undocumented)

updateRddInfo

public static void updateRddInfo(scala.collection.Seq<RDDInfo> rddInfos,
                 scala.collection.Seq<StorageStatus> statuses)

Update the given list of RDDInfo with the given list of storage statuses. This method overwrites the old values stored in the RDDInfo's.

Parameters:

rddInfos - (undocumented)

statuses - (undocumented)

getRddBlockLocations

public static scala.collection.Map<BlockId,scala.collection.Seq<String>> getRddBlockLocations(int rddId,
                                                                              scala.collection.Seq<StorageStatus> statuses)

Return a mapping from block ID to its locations for each block that belongs to the given RDD.

Parameters:

rddId - (undocumented)

statuses - (undocumented)

Returns:

(undocumented)