Storm Blueprints: Patterns for Distributed Real-time Computation

Implementing the architecture

Let's now delve into the details of the implementation. For example purposes, the following code assumes the topology is running locally. We use an in-memory queue instead of a persistent queue, and a hash map as our storage mechanism. In a real production implementation, we would most likely use a durable queuing system such as Kafka and a distributed storage mechanism such as Cassandra.

The data model

We will look at each of the topologies in depth, but first, let's have a look at the data model. To simplify things, we've encapsulated the game logic and the data model into two classes: Board and GameState.

The following is a listing of the Board class:

public class Board implements Serializable { public static final ...