Designing Event-Driven Architecture

A Comprehensive Guide

De: Shane Larson

Narrado por: Virtual Voice

Every enterprise is being told to go event-driven. Kafka is on every architecture diagram, “real-time” is in every executive deck — and a lot of what’s being built is expensive confusion.

Event sourcing adopted because a conference talk made it sound inevitable. Exactly-once delivery promised by people who have never operated a broker under load. CQRS bolted onto systems that needed a database index. This book is the corrective: the honest, comprehensive, design-level guide to event-driven architecture that actually works in production.

From the author of Designing Solutions Architecture for Enterprise Integration, it covers the fundamentals with precision — the real difference between events, messages, and commands — then goes where the blog posts stop: sagas and distributed transactions, schema evolution across forty consumers, idempotency and the exactly-once myth, and observability for systems where “the request” no longer exists. A dedicated chapter covers event-driven architecture in regulated industries — audit, retention, provable ordering — drawn from the author’s experience running integration platforms at a financial institution handling hundreds of millions of calls annually. And because the next generation of enterprise systems will be agentic, it shows how events become the nervous system of AI agent architectures.

Vendor-neutral throughout: you’ll learn to evaluate any broker or streaming service on its architectural properties — so the book doesn’t expire with the next major release.

What you'll learn:

• Events vs. messages vs. commands — and why conflating them corrupts designs
• The distributed log as an architectural primitive: partitions, ordering truths, and folklore
• Why exactly-once delivery is a myth — and how idempotency and the outbox pattern actually solve it
• Event sourcing and CQRS without the cult: when each earns its complexity, and when to say no
• Sagas, compensating actions, and a payment flow that fails halfway — walked step by step
• Schema evolution, contracts, and observability for asynchronous systems
• Event-driven architecture that survives an audit — retention, privacy, crypto-shredding, provable ordering
• Legacy integration: CDC, anti-corruption layers, and living with the mainframe
• Events as the backbone of AI agent systems — coordination, memory, and kill switches

This book is for you if:

• You’re an architect or tech lead designing — or defending — event-driven systems
• You need vendor-neutral judgment, not another Kafka tutorial
• You work in banking, insurance, or healthcare, where the auditors are not hypothetical
• You read Designing Solutions Architecture for Enterprise Integration and want the next volume

Events are facts. Contracts are forever. Design accordingly — start reading today.