Make Distributed Event Driven Easier !!!

EasyEvent

中文版本

Note: This project has not been published to the Maven central repository. You need to manually add it to your local or private repository for use.

Welcome to Star (Welcome Star)!!!

Home Page
GitHub

Introduction

What is EasyEvent

EasyEvent is a distributed, extensible, and eventually consistent real-time EventBus solution based on Google Guava's EventBus.
It mainly addresses issues such as lack of persistence and distributed machine scheduling in the Google EventBus.

Features

• Supports execution of synchronous events, asynchronous events, serial events/parallel events triggering.
• Supports event exception compensation and warning functions.

Problems Solved

1. Solves data consistency problems in event-driven architectures.
2. Addresses service instability caused by concentrated event triggers.
3. Resolves distributed event traceability issues.
4. Handles monitoring and early-warning for abnormal event executions.
5. Addresses event-driven challenges in Domain-Driven Design (DDD).
6. Solves various distributed event issues: e.g., excessive storage demands for events, sharp traffic spikes due to clustered events, slow processing blocking other normal event subscriptions, etc.

Software Architecture

To address these issues, EasyEvent abstracts several core components. It uses EventStorage, which can be extended via SPI implementations.
For distributed event scheduling, it utilizes EventTransfer, also extendable through custom SPI implementations.
The asynchronous event processing workflow is illustrated below:

Diagram for Single-node EasyEvent Processing:

Diagram for SOAEvent Processing:

Code Structure

• easyevent-common: Public module service

• easyevent-core: Core EventBus logic

• easyevent-storage: Storage Service

  • easyevent-storage-api: Storage Service API
  • easyevent-storage-jdbc: JDBC-based storage implementation

• easyevent-transfer: Event Transfer Service

  • easyevent-transfer-api: Event transfer protocol
  • easyevent-transfer-disruptor: Disruptor as an event transfer implementation
  • easyevent-transfer-rocketmq: RocketMQ as an event transfer implementation
  • easyevent-transfer-kafka: Kafka as an event transfer implementation

• easyevent-spring-boot-starter: EasyEvent Starter Package

  • easyevent-spring-boot-starter-parent: easyevent starter parent project
  • easyevent-spring-boot-starter-disruptor: Spring Boot starter with Disruptor as the event transfer implementation
  • easyevent-spring-boot-starter-kafka: Spring Boot starter with Kafka as the event transfer implementation
  • easyevent-spring-boot-starter-rocketmq: Spring Boot starter with RocketMQ as the event transfer implementation

• easyevent-spring-boot-starter-soa: EasyEvent SOA Support Starter Package

  • easyevent-spring-boot-starter-soa-api: easyevent soa event api project
  • easyevent-spring-boot-starter-soa-parent: easyevent soa starter parent project
  • easyevent-spring-boot-starter-soa-rocketmq: SOA-based event transfer center using RocketMQ as the implementation

• easyevent-example: Example Projects

  • easyevent-example-disruptor: Uses Disruptor as the transport scheduling layer
  • easyevent-example-rocketmq: Uses RocketMQ as the transport scheduling layer
  • easyevent-example-kafka: Uses Kafka as the transport scheduling layer
  • easyevent-example-soa: SOA Example Project
    • easyevent-example-soa-event: SOA public events
    • easyevent-example-soa-publisher: SOA event publishing service
    • easyevent-example-soa-subscriber: SOA event subscription service

Quick Start

Quick Start Guide

Extensibility

Extending EasyEvent

Comparison with Similar Frameworks

There are many event-driven frameworks available, including Google EventBus, SpringCloud-Bus, killbill-queue, and RocketMQ EventBridge.
Below is a comparison focused on each framework's key features:

	easy-event	Google EventBus	killbill-queue	SpringCloud-Bus	RocketMQ EventBridge
Description	A distributed, eventually consistent EventBus solution based on Google Guava, with similar usage	In-memory EventBus based on single machine	Extended based on DB + Google EventBus	Provides both standalone and distributed EventBus for SpringCloud	High-reliability, low-coupling, high-performance event-driven architecture based on RocketMQ 5.0
Distributed Support	Supports distributed environments across multiple machines in a single service, SOA microservices	Single-machine only	Supports distributed environments for a single service	Supports distributed environments in microservices	Supports distributed environments in microservices
Reliability	Uses local message tables to ensure no data loss in distributed event-driven scenarios, achieving eventual consistency	Events not persisted, potential for loss	Uses local message tables for distributed event-driven handling	Not persistent; reliability depends on MQ in distributed mode	Depends on the stability of RocketMQ
Ease of Use	Provides a SpringBootStarter package; simply introduce dependencies for quick adoption. Usage is nearly identical to Google EventBus	Simple to use	Same usage as Google EventBus	Usage similar to Spring Application Event	Introduce EventBridge-related dependencies
Operations & Maintenance	Common component ecosystem, no additional deployment required (DB + MQ)	No additional deployment needed	Depends on Oracle (no MySQL version available)	Common component ecosystem, no additional deployment required (depends on MQ)	Deployment requires both RocketMQ EventBridge and RocketMQ
Isolation	Supports customizable isolation (storage, transmission) for different events with extensibility	No isolation	No isolation	No isolation	No isolation
Stability	Asynchronous event triggering supports custom rate limiting for both consumption and sending; tag-level isolation supported	No rate control	Timed scheduling, difficult to control frequency, poor real-time performance	In distributed environments, heavily dependent on MQ stability; sudden spikes in events can easily cause bottlenecks	Depends on the stability of RocketMQ EventBridge and RocketMQ
Extensibility	Storage supports relational databases like JDBC, and can be customized via APIs. Transmission supports RocketMQ, Kafka, and can also be extended	Not extensible	Depends on Oracle; other storage components require code rewrites	Multiple transmission components provided, easy to extend	Strongly dependent on RocketMQ; no alternative components currently available

🌟Misc