Reactive Programming with Spring & Kotlin: Server-side Power Unleashed

This presentation aims to delve into the realm of Reactive Programming, elucidating how it can be harnessed by combining it with Spring and Kotlin. We will begin by elucidating the foundational concepts of Reactive Programming and expound on its paramount significance in the realm of modern web applications.

We will demonstrate how you can effectively implement Reactive Programming by leveraging the robust capabilities of Spring Framework and the versatility of Kotlin. You will gain insights into employing tools like WebFlux, Flow, and Coroutines to bolster the responsiveness and scalability of your server-side applications.

Additionally, we will take a deep dive into the merits of Server-Sent Events (SSE) and provide a comprehensive exploration of how to harness them for real-time data communication and updates.

This presentation is designed to be a comprehensive guide, bridging the gap between theory and practical application, to empower you with the knowledge and skills needed to fortify your server-side applications using Reactive Programming with Spring and Kotlin.

Transcript

1. Reactive Programming with Spring & Kotlin: Server-side Power Unleashed Gökhan Ayrancıoğlu
2. Gökhan Ayrancıoğlu Team Lead @Colendi programming, since 2009 gokhana.medium.com gokhanadev gokhana@mail.com gokhana.dev G-khan gokhanadev
3. Kotlin is a great fit for developing server-side applications. Full compatibility with existing Java-based technology Server-side Kotlin?
4. Reactive programming is a software paradigm based on asynchronous data streams and event-driven, real-time operations. Reactive Programming
5. Reactive Terminology Let's not block threads. Non-Blocking Everything is a stream in Reactive. Data Stream Processing based on load Backpressure Anything that can be asynchronous should be asynchronous. Async
6. Imperative Programming Due to blocking, only as many requests can be processed as the capacity of the thread pool allows.
7. Imperative Programming
8. Is Imperative Programming dying? No, it is not dying.
9. ★ Async and non-blocking. ★ Creates data flow through event-driven streams ★ Reactive programming != functional programming ★ There's backpressure in data streams ★ Predictable response times ★ Uses system resources much more efficiently. Reactive Programming Depths
10. — DAVE SYER Reactive is that you can do more with less, specifically you can process higher loads with fewer threads.
11. How does Reactive Programming work?
12. How does Reactive Programming Works ➔ The core of reactive programming is managing data through streams in an event-based way. ➔ When developing an application, you should create data flows for everything. ➔ An event or message is created for each result received from the data source. ➔ When a query is made to the database, the service immediately returns and pushes the data through the stream when it is ready. An event is created for each data item, and the data stream is completed with an onComplete.
13. Reactive Programming
14. What happens when an error occurs?
15. Reactive Programming
16. And Backpressure "If more load comes than we expect or can handle, we may not be able to process it. In case of a sudden increase in requests, there may be server or client crashes.
17. Tell me about Benchmark. Resource: https://itembase.com/resources/blog/tech/reactive-programming-performance-and-trade-offs/ • Blocking app: default number of 200 threads to serve the requests • Reactive app: 1 thread for the server and $CPU_CORES for worker threads, quad-core processor = 4 worker threads to serve requests The Good
18. The BAD, The UGLY • Learning difficulty • Debugging is not very easy • It's not easy to Keep it Simple(KISS) with Reactive programming • Great power, requires great attention. • Testing is not very easy either • Memory intensive •
19. Which applications are suitable for it? Her pardigmanın daha iyi çalıştığı alanlar vardır.
20. Types of Software Reactive Programming Can Be Applied to Monitoring, Logging Apps Games Social Media Apps Batch Apps IoT Apps
21. A few Important Points: The Reactive manifesto (see: reactivemanifesto.org) • Responsive. • Resilient. • Elastic. • Message driven. Rx - Reactive Extensions: (see: http://reactivex.io/) • Java: RxJava — JavaScript: RxJS — C#: Rx.NET,(Unity): UniRx • Scala: RxScala — Clojure: RxClojure — C++: RxCpp — Lua: RxLua • Go: RxGo — Ruby: Rx.rb — Python: RxPY — Groovy: RxGroovy • Kotlin: RxKotlin — JRuby: RxJRuby — Swift: RxSwift • PHP: RxPHP — Elixir: reaxive — Dart: RxDart …
22. Reactive Programming Mono Flux More and more.. Understanding and explaining reactive programming is much harder than applying it :)
23. Reactive Controller JDBC Database Driver Blocking Rest Calls Reactive service layer
24. Kotlin – Reactive Programming ➔ Kotlin core reactive support ➔ Reactive frameworks support (Spring Webflux, Ktor, Quarkus etc.) ➔ Coroutines feature for nonblocking and async processes ➔ Flow API for reactivity with data streams and supports backpressure ➔ Compatible with JVM improvements ➔ Integration with Reactive Extensions (RxJava/RxKotlin)
25. Spring Webflux and Reactive Programming
26. Mono is used for Publishers that can contain 0 or 1 event. Flux is used for Publishers that can contain 0..N events. FLUX Mono Spring Webflux Publishers
27. Controller Layer Annotation-Based Endpoint Typical MVC structure
28. Annotation-based Controller just like a normal REST service.
29. Controller Layer Functional Endpoint HandlerFunction is a interface for handling "ServerRequest" and "ServerResponse" Mono. RouterFunctions for routing
30. Router functions handler function
31. Spring WebClient: New generation reactive HTTP Client Reactive HTTP client for making non-blocking and asynchronous.
32. WebClient A sample request using Webclient RestTemplate is Dead, long live Webclient! - RestClient is coming soon for sync operations.
33. R2DBC - Reactive Relational Database Reactiveway for relational databases. - Spring R2DBC is not a full ORM like JPA - We can't make relationships like OneToMany directly.
34. Kotlin Coroutines Non-blocking, asynchronous code in a more sequential and readable manner. Cooperation [Routines]
35. Kotlin Coroutines Coroutines use a lightweight threading (not a thread) model, enabling you to pause and resume tasks without blocking threads, making it efficient for handling concurrency and parallelism. Source: https://kotlinlang.org/
36. In a CoroutineScope, an async operation is performed for a suspend function. Firstly, even numbers are processed. Suspend function for coroutine usage.
37. Coroutines has support for repositories from R2DBC
38. Reactive Kotlin: Flow API A way to handle asynchronous data streams and reactive programming within backpressure support.
39. Build a flow Shared flow with map Special pancakes using transform and filter Consuming flow within collect
40. every single operator will log detailed enable debug mode for that section of the strea consumes error observes all streams signals and traces them into logs
41. Server-Sent Events It is a technology that enables the sending of messages/notifications/events from the server to the client through an HTTP connection.
42. Server-sent events (SSE) List of pre-defined SSE field names include: • event: Event type • data: Event or message object • retry: Field specifying the time to establish a connection after a connection is lost • id: An assigned id for each message
44. CREDITS: This presentation template was created by Slidesgo, including icons by Flaticon, and infographics & images by Freepik Thank You gokhana.medium.com gokhanadev gokhana@mail.com gokhana.dev G-khan gokhanadev Gökhan Ayrancıoğlu