Patterns

35 items

System Design Pattern

API & Gatewaybffbackend-for-frontendapiclient-specificaggregationintermediate

Backend for Frontend Pattern

Client-specific API layers for different frontends

Used in: GraphQL, Mobile Apps, Web Apps|20 min read

Summary

Backend for Frontend (BFF) creates dedicated backend services for each client type (web, mobile, smart TV). Instead of one generic API serving all clients, each BFF is tailored to its client's specific needs - aggregating data, transforming responses, and handling client-specific logic. This enables optimal API design per platform, independent deployment, and better separation of concerns. Netflix uses BFFs for different devices, Spotify has mobile and web BFFs. The pattern shines when clients have vastly different needs.

Key Takeaways

Client-Optimized APIs

Mobile needs less data (bandwidth), web can handle more. BFF returns exactly what each client needs, no over-fetching or under-fetching.

Independent Evolution

Mobile team owns mobile BFF, can iterate independently. Web changes don't affect mobile and vice versa. Faster development velocity.

Client-Specific Logic

Mobile offline sync, web real-time updates, TV simplified navigation - each BFF handles platform-specific requirements without polluting core services.

Backend for Frontend Pattern

Why separate BFFs?

Mobile BFF: - Minimal payload (bandwidth costs) - Aggregated endpoints (reduce round trips) - Offline sync support - Push notification handling

Web BFF: - Rich data for powerful UI - Real-time WebSocket support - SEO considerations - Server-side rendering support

TV BFF: - Simplified navigation data - Large media URLs - Voice command support - Low-power device optimization

Summary

Key Takeaways

Client-Optimized APIs

Mobile needs less data (bandwidth), web can handle more. BFF returns exactly what each client needs, no over-fetching or under-fetching.

Independent Evolution

Mobile team owns mobile BFF, can iterate independently. Web changes don't affect mobile and vice versa. Faster development velocity.

Client-Specific Logic

Mobile offline sync, web real-time updates, TV simplified navigation - each BFF handles platform-specific requirements without polluting core services.

Avoid Generic API Trap

Generic APIs try to serve everyone, end up serving no one well. BFF embraces different needs explicitly.

Duplication Trade-off

Some logic duplicated across BFFs. Accept this for flexibility, or extract shared logic to common libraries.

Team Structure Alignment

BFF aligns with Conway's Law - mobile team builds mobile BFF, web team builds web BFF. Clear ownership.

Pattern Details

Backend for Frontend Pattern

Why separate BFFs?

Mobile BFF: - Minimal payload (bandwidth costs) - Aggregated endpoints (reduce round trips) - Offline sync support - Push notification handling

Web BFF: - Rich data for powerful UI - Real-time WebSocket support - SEO considerations - Server-side rendering support

TV BFF: - Simplified navigation data - Large media URLs - Voice command support - Low-power device optimization

Trade-offs

Aspect	Advantage	Disadvantage

Patterns

Horizontal Scaling Pattern

Retry with Backoff Pattern

Replication Pattern

Caching Strategies Pattern

Persistent Connections Pattern

Load Balancing Pattern

Fan-out Pattern

Fan-in Pattern

Circuit Breaker Pattern

Eventual Consistency Pattern

Queue-based Load Leveling Pattern

Bloom Filters Pattern

Time-Series Storage Pattern

Bulkhead Pattern

Batch Processing Pattern

Write-Ahead Log Pattern

API Gateway Pattern