Rate Limiter in Go

A simple and extensible rate limiting library in Go, implementing Token Bucket and Leaky Bucket algorithms. This project includes:

• Modular rate limiter implementations following SOLID principles
• A configurable simulation harness to test limiter behavior under load
• Runtime configuration via YAML or environment variables
• CLI flag support for quick runtime overrides

---

Token Bucket

Leaky Bucket

Features

• Token Bucket and Leaky Bucket rate limiting algorithms
• Configurable capacity, refill/leak rates, concurrency, and request load
• Simulation of realistic request patterns to benchmark limiter performance
• Easy configuration through config.yaml or environment variables (.env)
• CLI flag overrides for quick testing and experimentation

---

Getting Started

Installation

Clone the repo:

git clone https://github.com/Prayag2003/rate-limiter-in-go.git
cd rate-limiter-in-go
go mod tidy
cp .env.example .env  

---

Configuration

Using config.yaml (preferred)

rate_limiter_type: token       # "token" or "leaky"
capacity: 200                 # Bucket capacity (max tokens or queue size)
refill_rate: 100              # Tokens/sec for Token Bucket
leak_rate: 100                # Requests/sec for Leaky Bucket
rps: 200                     # Requests per second to simulate
duration_sec: 10             # Simulation duration in seconds
concurrency: 40              # Number of concurrent workers generating requests

Using environment variables (fallback)

Alternatively, set the following environment variables or define a .env file:

RATE_LIMITER_TYPE=token
CAPACITY=200
REFILL_RATE=100
LEAK_RATE=100
RPS=200
DURATION_SEC=10
CONCURRENCY=40

---

Running the Simulation

go run main.go

By default, it loads config from config.yaml if present; otherwise, from environment variables.

---

CLI Flag Overrides

You can override the rate limiter type at runtime with:

go run main.go --type=leaky

---

Code Overview

main.go

• Loads configuration (YAML or ENV)
• Parses CLI flags for runtime overrides
• Instantiates the requested limiter (Token Bucket or Leaky Bucket)
• Runs the simulation with configured RPS, duration, and concurrency

---

internal/limiter/

Contains:

• RateLimiter interface with method Allow() bool
• TokenBucket implementation (capacity, refill rate)
• LeakyBucket implementation (capacity, leak rate)

---

internal/simulator/simulator.go

Simulates concurrent requests at a specified RPS and duration:

• Spawns N workers (concurrency)
• Each worker sends requests at evenly spaced intervals
• Requests are checked against the rate limiter’s Allow() method
• Counts allowed vs rejected requests
• Prints a summary at the end:

[Simulation] RPS: 200 | Duration: 10s | Allowed: 950 | Rejected: 1050 | Time: 10.01s

---

Example Output

$ go run main.go --type=token
[CLI] Overriding limiter type to: token
Loaded configuration from config.yaml
[Simulation] RPS: 200 | Duration: 10s | Allowed: 980 | Rejected: 1020 | Time: 10.03s