Back to Library
gcp
System Design
intermediate
Social media content feed

News Feed System

System Design Classic

A news feed system generates personalized content streams for each user by aggregating posts from their connections and ranking them by relevance. The core challenge is the fan-out problem: when a user with millions of followers posts, how do you update millions of feeds efficiently? Recommended for engineers building social platforms where personalized content delivery and fan-out scalability are critical to user engagement.

Data Flow

Feed API
Post Events
Feed Service
Ranking Service
Fan-out Worker
Feed Cache (Memorystore)
Social Graph
Fan-out Topic

Share this architecture with your network

Service Breakdown (8 services)

Other8 services
Feed API
  • •Routes API traffic and enforces policies
  • •Manages authentication and rate limiting
  • •Provides a unified API endpoint
Feed Service
  • •Assembles photo feeds from followed accounts
  • •Ranks posts by engagement and recency signals
  • •Supports infinite scroll with cursor-based pagination
Ranking Service
  • •Scores content by engagement, recency, and affinity
  • •Uses ML models for personalized feed ordering
  • •Balances freshness with predicted user interest
Feed Cache (Memorystore)
  • •Caches data in-memory with sub-millisecond latency
  • •Supports Redis protocol for broad compatibility
  • •Scales vertically without downtime
Social Graph
  • •Stores follower and following relationships per user
  • •Supports efficient fan-out lookups by follower count
  • •Scales with automatic sharding across partitions
Post Events
  • •Delivers messages between decoupled services reliably
  • •Supports millions of messages per second
  • •Guarantees at-least-once delivery to all subscribers
Fan-out Worker
  • •Runs event-driven code without servers
  • •Scales instantly from zero to peak load
  • •Cost-effective for sporadic workloads
Fan-out Topic
  • •Delivers messages between decoupled services reliably
  • •Supports millions of messages per second
  • •Guarantees at-least-once delivery to all subscribers

Scaling Strategy

Feed generation uses a hybrid fan-out: posts from regular users are pre-computed into follower feeds stored in Memorystore, while posts from high-follower users are merged at read time. Pub/Sub handles the real-time event stream for feed updates. The ranking service runs on Cloud Run with ML-based scoring models. Firestore stores the social graph with subcollection patterns for efficient follower lookups.

Related Architectures

Multi-Tenant SaaS Platform

Generic SaaS

intermediate
System Design

Notification System

System Design Classic

intermediate
System Design

Dropbox File Storage System

Dropbox

intermediate
System Design

Chat Application System

System Design Classic

intermediate
System Design

Search Autocomplete System

System Design Classic

intermediate
System Design

Instagram Photo Sharing System

Instagram / Meta

intermediate
System Design