Java goes reactive

Reactive Streams in Java 9 with RxJava2

Transcript

1. > < JAVA GOES REACTIVE REACTIVE STREAMS IN JAVA 9 WITH RXJAVA2 1 KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI 13.06.2017
2. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Manifesto 2 Responsive Elastic Message driven Resilient
3. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Systems 
 vs Reactive Programming 3 Reactive systems • set of design rules • interaction between
 services • fulfilling Reactive
 Manifesto Reactive programming • asynchronous • non-blocking execution • event-driven
4. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Why Reactive Programming? 4 source: https://www.lightbend.com/reactive-programming-versus-reactive-systems
5. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxNetty performance 5 source: Reactive Programming with RxJava by Ben Christensen, Tomasz Nurkiewicz
6. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxNetty performance 6 source: Reactive Programming with RxJava by Ben Christensen, Tomasz Nurkiewicz
7. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Streams 7 “Reactive Streams is an initiative to provide a standard for asynchronous stream processing with non-blocking 
 back-pressure.” reactive-streams.org
8. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI What is a stream? 8 • series of items emitted over time
 • possibly no beginning, no end
 • used for transforming data
9. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Streams - what for? 9 • processing large data sets
 • real time data
 • monitoring and analytics
10. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Streams - how? 10 Publisher Subscriber Processor
11. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 11 Publisher Subscriber fast
 100 ops / s slow
 1 ops / s PUSH model Publisher passes item to Subscriber
12. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 12 Publisher Subscriber fast
 100 ops / s slow
 1 ops / s PUSH model Buffer Overflow
 Out Of Memory Publisher passes item to Subscriber
13. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 13 Publisher Subscriber slow
 1 ops / s fast
 100 ops / s PUSH model Publisher passes item to Subscriber
14. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 14 Publisher Subscriber slow
 1 ops / s fast
 100 ops / s PUSH model ✓OK Publisher passes item to Subscriber
15. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 15 Publisher Subscriber fast
 100 ops / s slow
 1 ops / s PULL model Subscriber gets items from Publisher 
 in blocking call
16. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 16 Publisher Subscriber fast
 100 ops / s slow
 1 ops / s PULL model ✓OK Subscriber gets items from Publisher 
 in blocking call
17. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 17 Publisher Subscriber slow
 1 ops / s fast
 100 ops / s PULL model Subscriber gets items from Publisher 
 in blocking call
18. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Back-pressure 18 Publisher Subscriber slow
 1 ops / s fast
 100 ops / s PULL model ✗ slow Subscriber gets items from Publisher 
 in blocking call
19. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Dynamic pull/push 19 Publisher Subscriber request (n) data (up to n)
20. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Dynamic pull/push 20 Publisher Subscriber request (n) data (up to n) fast subscriber => request(Long.MaxValue) => PUSH model
21. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Streams SPI 21 public interface Publisher<T> { public void subscribe(Subscriber<? super T> s); } public interface Subscription { public void request(long n); public void cancel(); } public interface Subscriber<T> { public void onSubscribe(Subscription s); public void onNext(T t); public void onError(Throwable t); public void onComplete(); } 
 public interface Processor<T, R> extends Subscriber<T>, Publisher<R> { }
22. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Streams Implementations 22 • Akka Streams • Project Reactor • RxJava / RxJava2 • MongoDB • Reactive Rabbit • Vert.x
23. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI 23 But we were supposed to talk 
 about Java 9, right?
24. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI Reactive Streams and JDK9 24 package org.reactivestreams; class java.util.concurrent.Flow; • JEP 266 • Java Flow API
25. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 25 “RxJava – Reactive Extensions for the JVM – a library for composing asynchronous and event-based programs using observable sequences for the Java VM.” https://github.com/ReactiveX/RxJava
26. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 26 • nice API (Java 8 Streams like)
 • transformations on streams
 • observer pattern on steroids
 • implements Reactive Streams SPI
27. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 27 Publisher Subscriber Operator Operator Operator
28. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 operators - flatMap 28
29. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 operators - take 29
30. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 operators - merge 30
31. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 operators - zip 31
32. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 operators 32 … and many, many more at http://rxmarbles.com
33. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI RxJava 2 example 33 Source code available at: https://github.com/krzysztof-pawlowski/TJUG-RxJava2
34. > < KRZYSZTOF PAWŁOWSKI | @KRZYCHPAWLOWSKI THANK YOU! QUESTIONS? 34