Reactive Systems in Java: Resilient, Event-Driven Architecture with Quarkus

Preface....7

Who Should Read This Book?....8

What About Quarkus?....9

Navigating This Book....9

Getting You Ready....11

Conventions Used in This Book....16

OReilly Online Learning....17

How to Contact Us....18

Acknowledgments....18

I. Reactive and Quarkus Introduction....20

1. Reactive in a Nutshell....21

What Do We Mean by Reactive?....21

Reactive Software Is Not New....22

The Reactive Landscape....23

Why Are Reactive Architectures So Well-Suited for Cloud Native Applications?....27

Reactive Is Not a Silver Bullet....27

2. Introduction to Quarkus....29

Java on the Cloud....29

Thorntail Example....34

Quarkus Example....37

The Quarkus Way....29

Create Your First Quarkus Application....29

Kubernetes with Quarkus in 10 Minutes....29

Going Native....59

Summary....64

II. Reactive and Event-Driven Applications....66

3. The Dark Side of Distributed Systems....67

Whats a Distributed System?....67

The New Kids on the Block: Cloud Native and Kubernetes Native Applications....67

The Dark Side of Distributed Systems....79

Fallacies of Distributed Computing in a Kubernetes World....80

A Question of Timing: The Synchronous Communication Drawback....83

Summary....91

4. Design Principles of Reactive Systems....93

Reactive Systems 101....93

Commands and Events....96

Commands....96

Events....97

Messages....97

Commands Versus Events: An Example....98

Destinations and Space Decoupling....102

Time Decoupling....105

The Role of Nonblocking InputOutput....93

Blocking Network IO, Threads, and Concurrency....107

How Does Nonblocking IO Work?....111

Reactor Pattern and Event Loop....93

Anatomy of Reactive Applications....116

Summary....121

5. Reactive Programming: Taming the Asynchronicity....122

Asynchronous Code and Patterns....122

Using Futures....129

Project Loom: Virtual Threads and Carrier Threads....132

Reactive Programming....122

Streams....137

Operators....139

Reactive Programming Libraries....145

Reactive Streams and the Need for Flow Control....145

Buffering Items....147

Dropping Items....148

What Is Backpressure?....149

Introducing Reactive Streams....149

Be Warned: Its a Trap!....153

Backpressure in Distributed Systems....153

Summary....154

III. Building Reactive Applications and Systems with Quarkus....156

6. Quarkus: Reactive Engine....157

The Imperative Model....157

The Reactive Model....161

Unification of Reactive and Imperative....157

A Reactive Engine....157

A Reactive Programming Model....171

Event-Driven Architecture with Quarkus....172

Summary....173

7. Mutiny: An Event-Driven Reactive Programming API....175

Why Another Reactive Programming Library?....175

What Makes Mutiny Unique?....177

Mutiny Usage in Quarkus....177

Uni and Multi....178

Mutiny and Flow Control....183

Observing Events....184

Transforming Events....185

Chaining Asynchronous Actions....186

Recovering from Failure....189

Combining and Joining Items....190

Selecting Items....192

Collecting Items....193

Summary....194

8. HTTP with Reactive in Mind....195

The Journey of an HTTP Request....196

Say Hello to RESTEasy Reactive!....198

Whats the Benefit?....201

Asynchronous Endpoints Returning Uni....204

Dealing with Failure and Customizing the Response....206

Streaming Data....210

Raw Streaming....210

Streaming JSON Array....212

Using Server-Sent-Events....215

Reactive Score....216

Summary....218

9. Accessing Data Reactively....220

The Problem with Data Access....220

Nonblocking Interactions with Relational Databases....223

Using a Reactive ORM: Hibernate Reactive....227

What About NoSQL?....231

Interacting with Redis....232

Data-Related Events and Change Data Capture....236

Using Debezium to Capture Change....239

Summary....243

IV. Connecting the Dots....245

10. Reactive Messaging: The Connective Tissue....246

From Reactive Applications to Reactive Systems....246

Channels and Messages....247

Producing Messages....248

Consuming Messages....252

Processing Messages....254

Acknowledgments....256

Connectors....258

Building Message-Based Applications....258

Message and Acknowledgment....260

Failures and Negative Acknowledgment....262

Stream Manipulation....263

Blocking Processing....265

Retrying Processing....266

Putting Everything Together....267

Summary....271

11. The Event Bus: The Backbone....272

Kafka or AMQP: Picking the Right Tool....272

Building Reactive Systems with Kafka....273

Apache Kafka....273

Point-to-Point Communication....280

PublishSubscribe....282

Elasticity Patterns....283

Dealing with Failures....284

Backpressure and Performance Considerations....287

Kafka on Kubernetes....272

Building Reactive Systems with AMQP....292

AMQP 1.0....292

Point-to-Point Communication....293

PublishSubscribe....295

Elasticity Patterns....296

Acknowledgment and Redelivery....297

Credit-Flow Backpressure Protocol....298

AMQP on Kubernetes....298

Summary....300

12. Reactive REST Client: Connecting with HTTP Endpoints....302

Interacting with an HTTP Endpoint....302

The REST Client Reactive....305

Mapping HTTP APIs to Java Interfaces....306

Invoking the Service....309

Blocking and Nonblocking....310

Handling Failures....311

Fallback....312

Retries....314

Time-out....315

Bulkheads and Circuit Breaker....315

Building API Gateways with the RESTEasy Reactive Client....317

Using the REST Client in Messaging Applications....324

Summary....328

13. Observing Reactive and Event-Driven Architectures....330

Why Is Observability Important?....330

Health with Messaging....332

Metrics with Messaging....339

Distributed Tracing with Messaging....344

Summary....348

Conclusion....350

A Brief Summary....350

Is That All?....351

The Future of Reactive Systems....352

The End of the Beginning....353

Index....354

Reactive systems and event-driven architecture are becoming indispensable to application design, and companies are taking note. Reactive systems ensure that applications are responsive, resilient, and elastic no matter what failures or errors may be occurring, while event-driven architecture offers a flexible and composable option for distributed systems. This practical book helps Java developers bring these approaches together using Quarkus 2.x, the Kubernetes-native Java framework.

Clement Escoffier and Ken Finnigan show you how to take advantage of event-driven and reactive principles to build robust distributed systems, reducing latency and increasing throughput, particularly in microservices and serverless applications. You'll also get a foundation in Quarkus to help you create true Kubernetes-native applications for the cloud.