Class SliceRangeShardAllocationStrategy
- java.lang.Object
-
- akka.cluster.sharding.typed.SliceRangeShardAllocationStrategy
-
- All Implemented Interfaces:
NoSerializationVerificationNeeded
,ClusterShardAllocationMixin
,ShardCoordinator.ActorSystemDependentAllocationStrategy
,ShardCoordinator.ShardAllocationStrategy
public class SliceRangeShardAllocationStrategy extends java.lang.Object implements ShardCoordinator.ActorSystemDependentAllocationStrategy, ClusterShardAllocationMixin
Intended to be used with database sharding, https://doc.akka.io/docs/akka-persistence-r2dbc/current/data-partition.html, with a cluster of many Akka nodes. To avoid that each Akka node has database connections to all databases it is preferred to collocate entities with the same slice and contiguous range of slices to the same Akka node. Thereby the connections from one Akka node will go to one or a few databases since the database sharding is based on slice ranges.It must be used together with the
SliceRangeShardAllocationStrategy.ShardBySliceMessageExtractor
.Create a new instance of this for each entity type, i.e. a
SliceRangeShardAllocationStrategy
instance must not be shared between different entity types.It will not rebalance when there is already an ongoing rebalance in progress.
Important: Do not change shard allocation strategy in a rolling update. The cluster must be fully stopped and then started again when changing to a different allocation strategy.
Not intended for public inheritance/implementation.
param: absoluteLimit the maximum number of shards that will be rebalanced in one rebalance round param: relativeLimit fraction (< 1.0) of total number of (known) shards that will be rebalanced in one rebalance round
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description static class
SliceRangeShardAllocationStrategy.ShardBySliceMessageExtractor<M>
To be used together withSliceRangeShardAllocationStrategy
.-
Nested classes/interfaces inherited from interface akka.cluster.sharding.internal.ClusterShardAllocationMixin
ClusterShardAllocationMixin.RegionEntry, ClusterShardAllocationMixin.RegionEntry$, ClusterShardAllocationMixin.ShardSuitabilityOrdering, ClusterShardAllocationMixin.ShardSuitabilityOrdering$
-
-
Constructor Summary
Constructors Constructor Description SliceRangeShardAllocationStrategy(int absoluteLimit, double relativeLimit)
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description scala.concurrent.Future<ActorRef>
allocateShard(ActorRef requester, java.lang.String shardId, scala.collection.immutable.Map<ActorRef,scala.collection.immutable.IndexedSeq<java.lang.String>> currentShardAllocations)
Invoked when the location of a new shard is to be decided.protected ClusterEvent.CurrentClusterState
clusterState()
scala.concurrent.Future<scala.collection.immutable.Set<java.lang.String>>
rebalance(scala.collection.immutable.Map<ActorRef,scala.collection.immutable.IndexedSeq<java.lang.String>> currentShardAllocations, scala.collection.immutable.Set<java.lang.String> rebalanceInProgress)
Invoked periodically to decide which shards to rebalance to another location.protected Member
selfMember()
void
start(ActorSystem system)
Called before any calls to allocate/rebalance.-
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
-
Methods inherited from interface akka.cluster.sharding.internal.ClusterShardAllocationMixin
isAGoodTimeToRebalance, regionEntriesFor
-
-
-
-
Method Detail
-
start
public void start(ActorSystem system)
Description copied from interface:ShardCoordinator.ActorSystemDependentAllocationStrategy
Called before any calls to allocate/rebalance. Do not block. If asynchronous actions are required they can be started here and delay the Futures returned by allocate/rebalance.- Specified by:
start
in interfaceShardCoordinator.ActorSystemDependentAllocationStrategy
-
clusterState
protected ClusterEvent.CurrentClusterState clusterState()
- Specified by:
clusterState
in interfaceClusterShardAllocationMixin
-
selfMember
protected Member selfMember()
- Specified by:
selfMember
in interfaceClusterShardAllocationMixin
-
allocateShard
public scala.concurrent.Future<ActorRef> allocateShard(ActorRef requester, java.lang.String shardId, scala.collection.immutable.Map<ActorRef,scala.collection.immutable.IndexedSeq<java.lang.String>> currentShardAllocations)
Description copied from interface:ShardCoordinator.ShardAllocationStrategy
Invoked when the location of a new shard is to be decided.- Specified by:
allocateShard
in interfaceShardCoordinator.ShardAllocationStrategy
- Parameters:
requester
- actor reference to theShardRegion
that requested the location of the shard, can be returned if preference should be given to the node where the shard was first accessedshardId
- the id of the shard to allocatecurrentShardAllocations
- all actor refs toShardRegion
and their current allocated shards, in the order they were allocated- Returns:
- a
Future
of the actor ref of theShardRegion
that is to be responsible for the shard, must be one of the references included in thecurrentShardAllocations
parameter
-
rebalance
public scala.concurrent.Future<scala.collection.immutable.Set<java.lang.String>> rebalance(scala.collection.immutable.Map<ActorRef,scala.collection.immutable.IndexedSeq<java.lang.String>> currentShardAllocations, scala.collection.immutable.Set<java.lang.String> rebalanceInProgress)
Description copied from interface:ShardCoordinator.ShardAllocationStrategy
Invoked periodically to decide which shards to rebalance to another location.- Specified by:
rebalance
in interfaceShardCoordinator.ShardAllocationStrategy
- Parameters:
currentShardAllocations
- all actor refs toShardRegion
and their current allocated shards, in the order they were allocatedrebalanceInProgress
- set of shards that are currently being rebalanced, i.e. you should not include these in the returned set- Returns:
- a
Future
of the shards to be migrated, may be empty to skip rebalance in this round
-
-