Revealed: The software scalability game | Samuel Lissner

🌈 Abstract

The article discusses the different levels of software scalability, from a monolithic architecture to a cloud-native, planet-scale architecture. It outlines the key characteristics and considerations at each level, and provides guidance on how to "level up" to the next stage of scalability.

🙋 Q&A

[01] The Monolith

1. What is a monolithic architecture? A monolithic architecture is a single deployable application that can handle up to hundreds of requests per second. By adding a load balancer in front of a few instances, it can serve thousands of requests per second.

2. When should a company start with a monolith? The article states that for most small and medium businesses, a monolithic architecture should already be good enough. A starting company should not even consider other forms of architecture unless its customer base grows.

3. What is the key advice regarding monoliths? The key advice is: DO NOT level up unless you really need to. The article states that you should almost always start with a monolith.

[02] The Modulith

1. What is a modulithic architecture? A modulithic architecture is a single deployable application composed of loosely coupled modules. It is easier to deploy and maintain than a monolith.

2. When is a modulith the way to go? The article states that for big companies or smaller enterprise applications, a modulith is the way to go.

3. How do you level up to a modulith? To level up to a modulith, you need to break up the monolith into separate domain modules that do not depend on each other. You can use tools like Gradle, the archunit library, or Spring Modulith (if using Spring) to enforce the boundaries between modules. Communication between modules can be done using Spring domain events or other event-based architecture.

[03] Microservices

1. What is a microservices architecture? Microservices consist of multiple deployable software assets that scale independently according to their needs. Each microservice should focus on a single domain and may have its own data storage.

2. What is the trade-off with microservices? While microservices can achieve superior scalability, the orchestration of multiple services comes at the cost of higher complexity.

3. When should you consider moving to microservices? The article advises that before even thinking about a microservices architecture, you should make sure you have reached the modulith level. You should have established clear domain boundaries, as it is much easier to break up a well-defined modulith than a messy, entangled monolith. You should really hit a wall with your modulithic scalability constraints before escalating to microservices.

[04] Cloud-Native Planet Scale

1. What is the "Cloud-Native Planet Scale" level? This level is where a few huge corporate companies, notably FAANG and other planet-scale companies, have reached a point where "everything scales".

2. How do you level up to this stage? To reach this level, you need to ensure that your entire application landscape is cloud-native. Every asset in the architecture should be scalable. The architect may need to shift to a data-driven approach to observe and manage the architecture.

3. What is the key takeaway regarding scalability levels? The article concludes that the choice of scale depends on your business needs. If you work in a FAANG-like company, "will it scale?" is a legitimate question. If you work in a startup without customers yet, it may be premature. The key is to optimize for adaptability rather than scalability, as it is much easier to level up understandable, scalable code than the other way around.