Architecture

Software design patterns, system design, and architectural decision-making.

Architecture is about making the right trade-offs. This section covers patterns, principles, and real-world case studies to help you design resilient, scalable systems.

Core Patterns

Separation of Concerns

Divide your system into distinct sections, each addressing a separate concern.

foundational

Layered Architecture

Presentation, business logic, and data access in isolated layers.

monolith · enterprise

Microservices

Independent services communicating over well-defined APIs.

distributed · cloud

Event-Driven Architecture

Services communicate through events — loose coupling, async processing.

async · scalable

Application Architectures

Phoenix LiveView Architecture

How endpoints, LiveViews, contexts, Ecto, PubSub, and OTP fit together into a real-time server-rendered application.

elixir · phoenix · liveview

Billing Management Architecture

Billing as a scheduling problem: metering, invoicing, dunning, and reconciliation orchestrated by a signed scheduled-webhook control plane.

billing · scheduling · webhooks

Case Studies

Seamonkeys Case Study

A non-proprietary case study to evaluate architectural concepts and framework decisions.

case study

Decision Framework

When choosing an architecture, consider:

Factor Monolith Microservices Event-Driven
Team sizeSmallMultiple teamsMultiple teams
Deploy frequencyLow–MediumHighHigh
ComplexityLowHighHigh
LatencyLowNetwork-dependentAsync
Data consistencyEasyHard (distributed tx)Eventual
ScalingVerticalHorizontal per serviceHorizontal per handler

Start simple. Extract services only when you have a clear reason — team boundaries, independent scaling, or deploy cadence.

A billing management architecture built around scheduled events — using Recurrence Cloud as the schedule control plane for invoicing, dunning, rollups, and reconciliation.

Services communicate through events for loose coupling and async processing.

Organize your application into presentation, business logic, and data access layers.

Independent services communicating over well-defined APIs.

Understand the architecture of a Phoenix LiveView application — how endpoints, routers, LiveViews, contexts, and Ecto fit together into a cohesive system.

Divide your system into distinct sections, each addressing a separate concern.