org.apache.flink.runtime.state

接口 SharedStateRegistry

所有超级接口:

AutoCloseable

所有已知实现类:

SharedStateRegistryImpl

public interface SharedStateRegistry
extends AutoCloseable

This registry manages state that is shared across (incremental) checkpoints, and is responsible for deleting shared state that is no longer used in any valid checkpoint.

A SharedStateRegistry will be deployed in the CheckpointCoordinator to keep track of usage of StreamStateHandles by a key that (logically) identifies them.

字段概要

字段

限定符和类型	字段和说明
static SharedStateRegistryFactory	DEFAULT_FACTORY A singleton object for the default implementation of a SharedStateRegistryFactory

方法概要

限定符和类型	方法和说明
void	checkpointCompleted(long checkpointId)
void	registerAll(Iterable<? extends CompositeStateHandle> stateHandles, long checkpointID) Register given shared states in the registry.
void	registerAllAfterRestored(CompletedCheckpoint checkpoint, RestoreMode mode) Set the lowest checkpoint ID below which no state is discarded, inclusive.
default StreamStateHandle	registerReference(SharedStateRegistryKey registrationKey, StreamStateHandle state, long checkpointID) Shortcut for registerReference(SharedStateRegistryKey, StreamStateHandle, long, boolean) with preventDiscardingCreatedCheckpoint = false.
StreamStateHandle	registerReference(SharedStateRegistryKey registrationKey, StreamStateHandle state, long checkpointID, boolean preventDiscardingCreatedCheckpoint) Register a reference to the given shared state in the registry.
Set<Long>	unregisterUnusedState(long lowestCheckpointID) Unregister state that is not referenced by the given checkpoint ID or any newer.

从接口继承的方法 java.lang.AutoCloseable

close

字段详细资料

DEFAULT_FACTORY

static final SharedStateRegistryFactory DEFAULT_FACTORY

A singleton object for the default implementation of a SharedStateRegistryFactory

方法详细资料

registerReference

default StreamStateHandle registerReference(SharedStateRegistryKey registrationKey,
                                            StreamStateHandle state,
                                            long checkpointID)

Shortcut for registerReference(SharedStateRegistryKey, StreamStateHandle, long, boolean) with preventDiscardingCreatedCheckpoint = false.

registerReference

StreamStateHandle registerReference(SharedStateRegistryKey registrationKey,
                                    StreamStateHandle state,
                                    long checkpointID,
                                    boolean preventDiscardingCreatedCheckpoint)

Register a reference to the given shared state in the registry. The registry key should be based on the physical identifier of the state. If there is already a state handle registered under the same key and the 'new' state is not equal to the old one, an exception will be thrown.

IMPORTANT: the caller must use the returned state handle instead of the passed one because the registry might replace or update it.

参数:

state - the shared state for which we register a reference.

checkpointID - which uses the state

preventDiscardingCreatedCheckpoint - as long as this state is still in use. The "checkpoint that created the state" is recorded on the first state registration.

返回:

the state handle registered under the given key. It might differ from the passed state handle, e.g. if it was a placeholder.

unregisterUnusedState

Set<Long> unregisterUnusedState(long lowestCheckpointID)

Unregister state that is not referenced by the given checkpoint ID or any newer.

参数:

lowestCheckpointID - which is still valid.

返回:

a set of checkpointID which is still in use.

registerAll

void registerAll(Iterable<? extends CompositeStateHandle> stateHandles,
                 long checkpointID)

Register given shared states in the registry.

NOTE: For state from checkpoints from other jobs or runs (i.e. after recovery), please use registerAllAfterRestored(CompletedCheckpoint, RestoreMode)

参数:

stateHandles - The shared states to register.

checkpointID - which uses the states.

registerAllAfterRestored

void registerAllAfterRestored(CompletedCheckpoint checkpoint,
                              RestoreMode mode)

Set the lowest checkpoint ID below which no state is discarded, inclusive.

After recovery from an incremental checkpoint, its state should NOT be discarded, even if not used anymore (unless recovering in CLAIM mode).

This should hold for both cases: when recovering from that initial checkpoint; and from any subsequent checkpoint derived from it.

checkpointCompleted

void checkpointCompleted(long checkpointId)