API Rate Limiter Microservice (Go)

A production-ready backend microservice for API rate limiting using the Token Bucket algorithm and Redis. Designed for high performance, clarity, and easy integration.

Features

• Token Bucket algorithm (allows bursts, smooths traffic)
• Global and per-client rate limits
• Redis for distributed, atomic state
• REST API: POST /check to verify if a request is allowed
• Configurable via environment variables
• Dockerized for local development
• Middleware for easy integration
• Unit tests for core logic and Redis

Why Token Bucket?

• Allows short bursts while enforcing a long-term average rate
• More flexible than fixed/sliding window for real-world APIs
• Efficient with Redis atomic operations

Architecture

Client → [POST /check] → Go API → RateLimiter (Token Bucket) → Redis

• Each client (by IP, API key, etc) has a token bucket in Redis
• On each request, the bucket is checked/updated atomically
• If tokens remain, request is allowed; else, denied with retry_after

Setup

Prerequisites

• Docker & Docker Compose

Run Locally

git clone <repo>
cd api-rate-limiter-microservice
docker-compose up --build

Environment Variables

• REDIS_URL (default: redis://localhost:6379)
• RATE_LIMIT (default: 100)
• WINDOW_SIZE (default: 1m)
• PORT (default: 8080)

API Usage

Check Rate Limit

POST /check
Content-Type: application/json
{
  "client_id": "user-123"
}

Response:

• Allowed: { "allowed": true }
• Rate limited: { "allowed": false, "retry_after": 42 }

Integration

• Use the provided Go middleware for easy plug-in to other services.

Testing

docker-compose exec api-rate-limiter go test ./...

Extending

• Add per-client custom limits in rateLimiter.go
• Add dashboard/CLI for usage stats (bonus)

License

MIT