Building Reactive Services Using Functional Programming
summary
Claims that F# lets services be message driven and resilient using concrete tools: immutable IEvent streams, async workflows with let! and async StartWithContinuations, and MailboxProcessors (agents). Demonstrations create 100,000 agents on a laptop, show an email-sending proof‑of‑concept with a supervisor agent that retries or logs failures, and explain that F# agents are in‑process only so remoting uses libraries like cricket or alternatives (Akka.NET, Orleans). Code examples include cancellation, reply channels, and elastic per‑URL agent spawning.
AI-written summary · about
chapters (14)
- 0:00Intro
- 1:59Taking on Amazon
- 3:25Reactive manifesto
- 4:12Events in F#
- 6:20Event combinators
- 8:38Computation expressions
- 10:11Consider classic synchronous code
- 12:43What is an actor?
- 14:34Agents: basic syntax
- 18:49One major difference from Erlang
- 22:04Error handling
- 24:56Supervisors
- 28:50Scaling agents on demand
- 30:01General resources & additional reading
related talks
33mRock Solid Haskell Services (Part 6)
47mWhirlwind Tour of the Scala Functional Programming Ecosystem
01h 02mMachine Learning with F#, Redux
53mREPL Driven Development, Scrum, and the Wright Brothers
59mRock Solid Haskell Services (Part 1)
43mThinking Functionally with C#
53mWhirlwind tour of the Scala Functional Programming ecosystem
01h 07mBuilding Reactive Pipelines