Implementing an event-driven microservices architecture in a functional language

Web services are typically stateless entities that need to operate at scale. Nikhil Barthwal explains how Jet.com implemented event-driven microservices architecture using a functional language, demonstrating how the functional paradigm captures the behavior of this service architecture very naturally and offers several benefits, such as scalability, productivity, and correctness.

Jet.com was launched in July 2015 with one mission: to become the smartest way to shop and save on pretty much anything. On average, the company receives over 25K orders per day and has over 15 million items in inventory. It has seen massive growth and was recently acquired by Walmart for $3.3 billion. Jet’s platform is implemented via microservices (currently 1,000+ running in production) and runs on Microsoft Cloud, using a large number of technologies and middleware. Microservices enable Jet to scale easily for independent releasability. Nikhil Barthwal explains how Jet.com implemented event-driven microservices architecture using F#, a statically typed, managed functional language based on the .NET platform, demonstrating how the functional paradigm captures the behavior of this service architecture very naturally. F# offers several benefits, such as scalability, productivity, and correctness. Immutability is important because it allows you to scale almost infinitely, because you never need to worry about a mutex, lock, or race. The productivity it provides is unmatched, as the code in F# is much more concise compared to any object-oriented language (typically 5x smaller). Its strict typing makes writing correct code easy, as the type checker infers the types for a new line of code and then double-checks that these types perfectly fit into the program. If there’s a mismatch of types, it is caught at compile type. The code also passes several rigorous mathematical tests of consistency, so it is much more likely to be correct.