Pattern · Distributed systems

Microservices Architecture

Independently deployable services, each owning its data and business capability.

Overview

Microservices decompose a monolith into small, independently deployable services. Each service owns its business domain and data store, communicates through APIs or events, and can be scaled, deployed, and failed in isolation.

Core Principles

Single Responsibility

Each service does one thing and does it well. Bounded by a business capability, not a technical layer.

Decentralized Data

Each service owns its own database. No shared schemas. Consistency through events, not transactions.

Fault Isolation

One service failing doesn't cascade. Circuit breakers, bulkheads, and timeouts contain failure.

Deploy Independently

Each service has its own CI/CD pipeline. Deploy without coordinating with other teams.

Key Patterns

API Gateway — single entry point, handles routing, auth, rate limiting
Service Discovery — services find each other dynamically (AWS Cloud Map)
Circuit Breaker — fail fast, prevent cascading failures
Saga Pattern — distributed transactions across services via events
Event Sourcing — store state as a sequence of events for full audit trail

AWS Services

Compute: ECS, EKS, Fargate, Lambda
Communication: API Gateway, SQS, SNS, EventBridge
Data: RDS (per service), DynamoDB, ElastiCache
Observability: X-Ray (distributed tracing), CloudWatch, OpenSearch

References

ECS Microservices CloudFormation

Complete architecture with ECS, ALB, and service discovery

GitHub

Terraform EKS Module

EKS cluster for Kubernetes-based microservices

GitHub

Microservices on AWS

AWS whitepaper — patterns, anti-patterns, and implementation guides

AWS Whitepaper

Well-Architected Microservices Lens

Framework guidance specifically for microservices

AWS Docs