Rate Limiter in Go

This repository demonstrates three different approaches to implementing rate limiting in Go:

1. Per-Client Rate Limiting (IP-Based)
2. Token Bucket Algorithm
3. Tollbooth Middleware

Each implementation is organized into separate folders for clarity and modularity.

---

1. Per-Client Rate Limiting (IP-Based)

This method limits requests per unique IP address. It maintains a map of clients and applies rate limiting per client using the golang.org/x/time/rate package.

Key Features:

• Tracks clients based on IP
• Uses rate.Limiter for rate limiting
• Cleans up inactive clients after 3 minutes

How to Run:

cd perclientlimit
go run main.go

---

2. Token Bucket Algorithm

A general rate-limiting technique that allows a burst of requests up to a limit and refills tokens at a fixed rate.

Key Features:

• Implements a token bucket system
• Uses rate.Limiter(2, 4), allowing 2 requests per second with a burst capacity of 4

How to Run:

cd tokenbucket
go run main.go

---

3. Tollbooth Middleware

This method uses the tollbooth package to implement rate limiting as middleware, making it easy to integrate with existing applications.

Key Features:

• Simple and effective middleware
• Customizable error messages and response types
• Limits requests using tollbooth.NewLimiter(1, nil), allowing 1 request per second

How to Run:

cd tollbooth
go run main.go

---

API Endpoints

Endpoint	Description	Rate Limit
/pingp	IP-based rate limiter	2 req/sec, burst 4
/ping	Token bucket rate limiter	2 req/sec, burst 4
/pingtoll	Tollbooth middleware limiter	1 req/sec

---

Installation & Dependencies

Ensure you have Go installed and run:

go mod tidy

Install Tollbooth if needed:

go get -u github.com/didip/tollbooth/v7

---

Author

Likhith K.P

License

MIT License