Memory Store

The MemoryStore is an in-memory implementation of the Store interface that provides per-session key-value storage backed by a thread-safe ConcurrentHashMap. It is lightweight, fast, and suitable for single-node deployments, development environments, and scenarios where session data does not need to be shared across multiple server instances.

Key characteristics

• Local per-session storage — each connection receives its own store instance

• Thread-safe access — supports concurrent reads/writes using ConcurrentHashMap

• Ephemeral data — values exist only in JVM memory and are lost on restart or failure

• No cross-node sharing — data is not distributed between servers

• Zero external dependencies — simplest possible Store implementation

How it works

• set(key, value) stores a value in memory scoped to the session

• get(key) retrieves the stored value, or null if none exists

• has(key) checks key presence without reading the value

• del(key) removes a single entry

• destroy() clears all entries, typically called when the client disconnects

Advantages

👍 Fast, low-latency access 👍 No external services required 👍 Ideal for development and local testing 👍 Works well with any EventStore for hybrid deployments (e.g., MemoryStore + KafkaEventStore or MemoryStore + RedisStreamEventStore)

Limitations

ℹ️ Not distributed — values are not synchronized across nodes ℹ️ Not persistent — data vanishes on server restart or crash ℹ️ Not suitable for horizontal scaling unless combined with a distributed StoreFactory implementation

---

Summary

MemoryStore provides local, ephemeral per-session data storage, ideal for single-node setups and development. Can be combined with any distributed EventStore when distribution of events is required without distributed storage.