Cloud Architecture
Production-grade cloud patterns for event-driven systems, serverless, resilience, data management, and deployment strategies.
24 conceptsInteractive diagrams
01
Strangler Fig Pattern
Incrementally replaces a legacy system by routing traffic to new services over time.
02
Sidecar Pattern
Deploys a helper container alongside each service to handle cross-cutting concerns.
03
Ambassador Pattern
Places a proxy container in front of a service to handle network tasks like retries and routing.
04
Anti-Corruption Layer (ACL)
Translates between two domain models to prevent a legacy system from polluting a new one.
05
Event-Driven Architecture
Decouples services by having them communicate through events rather than direct calls.
06
Event Sourcing
Stores state as an immutable sequence of events rather than current values.
07
CQRS — Command Query Responsibility Segregation
Separates the write model from the read model for independent scaling and optimisation.
08
Saga Pattern
Manages distributed transactions across services using a sequence of compensating actions.
09
Outbox Pattern
Reliably publishes events by writing them to a database outbox before sending.
10
Serverless Architecture
Runs code in ephemeral, fully-managed functions triggered by events.
11
Fan-Out / Fan-In Pattern
Splits work across parallel workers then aggregates results back together.
12
Competing Consumers Pattern
Multiple workers pull from the same queue to process messages concurrently.
13
Durable Execution Pattern
Persists workflow state so long-running processes survive crashes and restarts.
14
Circuit Breaker Pattern
Stops calling a failing service to give it time to recover and prevent cascade failures.
15
Bulkhead Pattern
Isolates failures by partitioning resources so one overloaded consumer cannot sink others.
16
Retry with Exponential Backoff & Jitter
Re-attempts failed calls with exponential backoff and randomised jitter.
17
Throttling & Rate Limiting Pattern
Controls the rate of outbound calls using token buckets or sliding windows.
18
Database per Service Pattern
Each microservice owns its own database to enforce loose coupling.
19
CQRS + Read Replicas Pattern
Combining separate write and read models with read replicas for maximum read throughput.
20
Materialized View Pattern
Pre-computes and caches query results for fast reads at the cost of staleness.
21
Blue/Green Deployment
Maintains two identical environments and switches traffic instantly for zero-downtime deploys.
22
Canary Deployment
Rolls out a new version to a small percentage of users before a full release.
23
Feature Flags Pattern
Decouples deployment from release, enabling runtime toggling of features.
24
Sidecar Service Mesh Pattern
Handles service-to-service communication, observability, and security via a dedicated infrastructure layer.