Cluster

As explained in Akka in Action, Second Edition

Akka Cluster is the Akka module that coordinates actors and runtime nodes into a distributed cluster.

A Cluster joins independent JVM-based ActorSystems into a peer-to-peer whole. Nodes can be added or removed as demand changes, while decentralized coordination avoids a single point of failure and supports fault tolerance. section 8.0

Cluster builds on akka-remote to make actor location transparent: business logic can communicate with an actor without needing to know whether it runs locally or on another node. Its core goals are fault-tolerant membership, distributing actor subtrees across nodes, and automated failover. section 8.2

Figure 8.1 A four-node clustered ActorSystem

Use the four-node view to distinguish one clustered ActorSystem deployment from a single JVM: each node participates in the same peer-to-peer cluster. section 8.2

A practical cluster setup separates formation and operations from application behavior:

• Configure the actor provider, remote settings, and seed nodes, then start an ActorSystem to form a basic cluster. section 8.6
• Track member lifecycle transitions: Joining, WeaklyUp, Up, Leaving, Down, Exiting, and Removed. section 8.6
• Use Akka Management and Management Cluster HTTP to inspect or change node state over HTTP; use Cluster Bootstrap and discovery instead of static contact configuration when appropriate. section 8.1
• Keep business logic independent of cluster formation so application behavior can be tested without cluster concerns. section 8.6

Resilient word counting

A word-counting workload can run actors on different nodes. After some nodes stop, new nodes can join and counting continues without lost work, illustrating how clustering combines scale-out processing with resilience. section 8.2 section 8.6

Cluster capabilities extend beyond membership. Cluster Singleton keeps one relocatable actor instance; Cluster Sharding distributes actors by partition key; Distributed Data supplies eventually consistent key-value objects; and Reliable Delivery adds at-least-once guarantees. section 8.1

How Akka in Action, Second Edition covers it

How deep it goes

Works through cluster formation, joining and leaving, member states, and notifications; defers the detailed word-count implementation until basics are established.

Examples

Uses a four-node ActorSystem diagram, a word-counting workload, and locally bootstrapped nodes with fixed IP configuration.

What it assumes

Builds from zero on cluster formation while treating ActorSystems as the nodes that participate.

What you can do after

Operate/configure it by setting provider, remote, and seed-node configuration, starting nodes, and handling membership changes.

How it connects

Connects akka-remote, location transparency, seed nodes, Cluster Bootstrap, Akka Management, Cluster Singleton, and Cluster Sharding.

liveBookmanning.com

Read Akka in Action, Second Edition on liveBook Get the book at manning.com

Related concepts

• gossip protocolused by · A gossip protocol is a distributed communication technique in which participants periodically exchange information with selected peers so state propagates through the network.
• horizontal scalingused by · Horizontal scaling is adjusting system capacity by adding or removing computing nodes or equivalent instances that share or distribute the workload.
• cluster shardingused by · Cluster sharding is an Akka technique that distributes actors across cluster nodes according to a partition key.
• cluster nodehas part · A cluster node is an individual physical or virtual machine that participates in a distributed cluster and provides computing resources or services.

This explainer was generated by Manning's knowledge graph from the text of Akka in Action, Second Edition. The section links point to the parts of the book it draws on.