Reactive Architecture <> Reactive Programming

12 Oct 2018

• Reactive Systems and Reactive Programming are often misunderstood.
• They are not equivalent.
• Reactive Systems apply the Reactive Principles at the architectural levels.
• Reactive Programming can be used to build Reactive Systems (or not).

Reactive Architecture

https://www.reactivemanifesto.org/

• The Reactive Manifesto principles are intrinsic to the design and architecture of Reactive Systems.
• All major architectural components interact in a Reactive way.
• Reactive Systems are separated along asynchronous boundaries.
• Eg. Reactive Microservices.

Reactive Principles

• Responsive: A Reactive System consistently responds in a timely fashion.
  • Responsiveness is the cornerstone of usability.
  • Systems must respond in a fast an consistent fashion if at al possible.
  • Responsive systems build user confidence.
• Resilient: A Reactive System remains responsive, even when failures occur.
  • Resilience provides responsiveness, despite failures.
  • Achieved through replication, isolation, containment, delegation.
  • Failures are isolated to a single Component.
  • Recovery is delegated to an external Component.
• Elastic: A Reactive System remains responsive, despite changes to system load.
  • Elasticity provides responsiveness, despite increases (or decreases) in load.
  • Implies zero scaling techniques can be used to support elasticity.
  • predictive auto scaling techniques can be used to support elasticity. scaling up provides responsiveness during peak, while scaling down improves cost effectiveness.
• Message Driven: A reactive System is built on foundation of async, non-blocking messages.
  • Responsiveness, Resilience, Elasticity are supported by a Message Driven Architecture.
  • Messages are asynchronous and non-blocking.
  • Provides Loose coupling, isolation, location transparency.
  • Resources are consumed only while active.

Reactive Programming

• Reactive Programming can be used to support the construction of Reactive Systems.
• Support breaking problems into small, discrete steps.
• Steps are executed in an async/non-blocking fashion, usually via a callback mechanism.
• Eg. Futures/Promises, Streams, RxJava/RxScala
• A system that uses Reactive Programming is not necessarily a Reactive System.

The Actor Model

• The Actor Model is a programming paradigm that supports construction of Reactive Systems.
• It is Message Driven.
• Abstractions provide Elasticity and Resilience.
• It can be used to build Responsive Software.

• All computation occurs inside of Actors.
• Each Actor has an address.
• Actors communicate only through asynchronous messages.
• The Message Driven nature of Actors supports Location Transparency.
• Actors communicate using the same technique, regardless of location.
• Local vs Remote is mostly configuration.
• Location Transparency enables actors t be both Resilient and Elastic.