Subatomic, Supersonic, Reactive
summary
Quercus claims to move framework work to build time so services start much faster and use less memory — author measures about 20% JVM startup/memory reduction and demo services running in ~54 MB with sub‑second startup. Demo wires MicroProfile Reactive Messaging annotations (@Channel + emitter, @Incoming/@Outgoing) to Kafka topics, uses SSE to push events to the browser, and shows durable ordering and automatic elasticity: orders queue when a consumer is down and multiple consumers share load when they restart.
AI-written summary · about
chapters (23)
- 0:00Start
- 3:44REACTIVE ALL THE THING
- 7:19Is Reactive == Event-Driven ?
- 8:01Event-Driven = Concurrent Applications
- 10:07TWO SIDES OF THE SAME COIN
- 10:48REACTIVE = RESPONSIVE
- 12:35The hidden truth of containers
- 16:00HTTP = STRONG COUPLING
- 16:54THE RISE OF CIRCUIT BREAKERS
- 18:14MESSAGING - 25+ YEARS OF EXPERIENCE
- 18:56DATA AT REST = DATA IN MOTION
- 20:12QUARKUS = (REACTIVE + IMPERATIVE) / REACTIVE
- 22:40REACTIVE COFFEE SHOP
- 24:30WHAT'S THE PROBLEM WITH SYNCHRONOUS
- 25:25DEMO
- 34:53DROP HTTP, SEND MESSAGES
- 35:46PROCESSING MESSAGES
- 36:04PUSHING DATA TO SSE
- 37:12MAPPING CHANNELS TO KAFKA TOPICS
- 38:09RUSH HOUR!
- 38:45QUARKUS = HTTP, MESSAGING, STREAMING
- 41:44WHAT ABOUT REACTIVE STREAMS?
- 42:42Supersonic, Subatomic, REACTIVE
related talks
02h 51mReactive Microservices with Quarkus
48mSupersonic, Subatomic Eclipse MicroProfile
49mSupersonic subatomic Enterprise Java from the ground up
45mProcessing 1m msgs/CPU/GB of memory with Quarkus and MicroProfile ...
50mBuilding reactive microservices with MicroProfile
28mReactive Quarkus
49mThe road to Quarkus. Why and what to look for
54mDeveloping a stream processing application with Apache Kafka and Quarkus