org.apache.flink.runtime.checkpoint

类 MappingBasedRepartitioner<T>

java.lang.Object

org.apache.flink.runtime.checkpoint.MappingBasedRepartitioner<T>

所有已实现的接口:

OperatorStateRepartitioner<T>

@NotThreadSafe
public class MappingBasedRepartitioner<T>
extends Object
implements OperatorStateRepartitioner<T>

A repartitioner that assigns the same channel state to multiple subtasks according to some mapping.

The replicated data will then be filtered before processing the record.

Note that channel mappings are cached for the same parallelism changes.

构造器概要

构造器

构造器和说明
MappingBasedRepartitioner(RescaleMappings newToOldSubtasksMapping)

方法概要

限定符和类型	方法和说明
List<List<T>>	repartitionState(List<List<T>> previousParallelSubtaskStates, int oldParallelism, int newParallelism)

从类继承的方法 java.lang.Object

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

构造器详细资料

MappingBasedRepartitioner

public MappingBasedRepartitioner(RescaleMappings newToOldSubtasksMapping)

方法详细资料

repartitionState

public List<List<T>> repartitionState(List<List<T>> previousParallelSubtaskStates,
                                      int oldParallelism,
                                      int newParallelism)

指定者:

repartitionState 在接口中 OperatorStateRepartitioner<T>

参数:

previousParallelSubtaskStates - List with one entry of state handles per parallel subtask of an operator, as they have been checkpointed.

oldParallelism - The parallelism before we start redistribution.

newParallelism - The parallelism that we consider for the state redistribution. Determines the size of the returned list.

返回:

List with one entry per parallel subtask. Each subtask receives now one collection of states that build of the new total state for this subtask.