Reactive programming for java developers
Download as PPTX, PDF0 likes79 views
This document provides an overview of reactive programming and how it can be applied to Java applications using frameworks like Project Reactor and Spring WebFlux. It defines reactive programming as composing asynchronous and event-based streams using non-blocking operators. It discusses how reactive streams like Flux and Mono work in Project Reactor and how Spring WebFlux allows building reactive web applications that can handle high concurrency with low latency. Examples are provided of building reactive data repositories, handling streaming responses with backpressure, and orchestrating remote service calls reactively.
Reactive programming for java developers
- 1. Reactive Programming for Java Developers Constantin PopaEngineering #5 16:00-16:45 Reactive Programming for Java Developers Constantin Popa
- 2. Reactive Programming for Java Developers Constantin Popa
- 4. Who here uses… Java 8 Spring framework RxJava/Reactor Spring WebFlux
- 5. AGENDA Reactive Programming 101 Reactive streams – Project Reactor Reactive web apps – Spring WebFlux
- 7. Why? because blocking is evil
- 8. sync&blocking main thread I/O time main thread resume processing
- 9. async&blocking main thread time wait & join I/O I/O
- 10. async&nonblocking main thread time Event loop worker thread I/O request I/O request I/O Data available
- 11. how do achieve that? without losing your mind…
- 13. “Composing asynchronous & event-based streams using non-blocking operators”
- 14. We don’t sacrifice much callbacks? Future? CompletableFuture?
- 15. Pull? Push!
- 17. Data in Flux
- 18. Publisher Subscriber produces consumes push events Feedback interfaces from Reactive Streams specification now part of Java 9, under java.util.concurrent.Flow Subscriber<T> onNext(T) onComplete() onError(Throwable) onSubscribe(Subscription) 0..N elements + 0..1 (complete | error) backpressure These interfaces don’t do much… Subscription.request(n) Subscription.cancel()
- 19. Project Reactor
- 21. Marble diagrams
- 22. Give the Flux a second chance
- 23. Interactive marble diagrams on http://rxmarbles.com (most map to Reactor operators)
- 26. Reactive streams vs Java 8 streams ”The design center for Streams is mostly about data that can be accessed with no latency (either from data structures or generating functions)” “The design center for Rx is about infinite streams of events which may arrive asynchronously.” Brian Goetz
- 27. an Rx-inspired API with a vocabulary of operators similar to RxJava
- 28. Flux.range map filter buffer Subscriber Nothing happens until you subscribe
- 33. Flux.range(1,3) .map( number -> toShape()) .filter(shape -> isCircle()) .buffer() .subscribe(shape -> printShape())
- 34. More awesomeness async sub-processes with flatMap
- 37. Flux.range(1,3) .map( number -> toShape()) .filter(shape -> isCircle()) .buffer() .subscribe(shape -> printShape())
- 38. Flux.range(1,3) .subscribeOn(Schedulers.elastic()) .map(number -> toShape()) .filter(shape -> isCircle()) .buffer() .publishOn(Schedulers.immediate()) .subscribe(shape -> printShape())
- 40. Reactive web applications Spring WebFlux
- 41. Use case #1 Reactive data repository
- 42. Let's recap: HTTP GET with reactive data repository Designed to work with both Spring MVC and Spring WebFlux Simply return reactive types(Flux, Observable) from @Controller @GetMapping("/cars") @ResponseBody public Flux<Car> getCars() { return this.repository.findAll(); }
- 43. Flux<T>: finite collection or infinite stream?
- 44. Use media type to decide
- 47. DEMO
- 48. Use case #2 Response stream with backpressure
- 50. DEMO
- 51. Let’s recap: HTTP GET with streaming response Simply return reactive types(Flux, Observable) from @Controller Backpressure handling on both Spring MVC and Spring WebFlux
- 52. Use case #3 Reactive remote service orchestration
- 53. DEMO
- 54. Let’s recap: Reactive webclient Orchestrate non-blocking, nested remote service calls with ease. Similar to reactive data access
- 57. What WebFlux is good for: High traffic High concurrency with less hardware resources Latency, streaming scenarios
- 58. Summary Quite a steep learning curve Doesn’t fit all scenarios, if it ain’t broken, don’t fix it It’s fun time to be a Java developer! Spring WebFlux and Reactor unlock the gates for easy Reactive Programming on the server side and not only