Explore stream-based programming and its potential to unify IO with functional programming in this 54-minute conference talk from GOTO Aarhus 2014. Dive into Akka Streams, an asynchronous library with non-blocking back pressure built on top of the Akka Actor Model. Learn about the Flow API, Flow Materialization, and the implications of non-blocking back pressure for fan-in and fan-out scenarios. Discover the concepts of collections and streams, user expectations, and unexpected behaviors in observed sequences. Examine possible solutions, including the Four Horsemen of Reactive and Reactive Streams. Understand the goals and motivations behind these technologies, and explore opportunities for API design, self-tuning back pressure, operation fusion and elision, execution optimizations, and distributed streams. Gain insights into Akka Actors and canonical papers in the field. By the end of the talk, acquire knowledge on how to get hands-on experience with these cutting-edge technologies.
How I Learned to Stop Worrying and Love the Flow
Overview
Syllabus
Introduction
What is a Stream?
What is a collection?
User Expectations
Unexpected: observed sequence depends on
java.util.stream
Streams vs. Collections
Possible Solutions
The Four Horsemen of Reactive
Origin and motivation
Collaboration between Engineers
Goals
Reactive Streams
A Data Market using Supply & Demand
Dynamic Push-Pull
Explicit Demand: One-to-many
Explicit Demand: Many-to-one
The Meat: Java
The dessert: Java
How does it Connect?
How does it Flow?
How does it Complete?
Akka Actors
Canonical papers
Opportunity: API
Opportunity: Self-tuning back pressure
Opportunity: Operation Fusion
Opportunity: Operation Elision
Opportunity: Execution optimizations
Opportunity: Distributed Streams
Outro: How do I get my hands on this?
Taught by
GOTO Conferences
