README.md

Rate Limiting

Distributed rate limiting algorithms for API throttling and traffic control.

Overview

Rate limiting protects services from overload by controlling request rates. This project implements token bucket, leaky bucket, sliding window, and distributed rate limiting algorithms.

Version: 0.1.0

Features

• ✅ Token Bucket — Burst support with refill
• ✅ Leaky Bucket — Smooth rate limiting
• ✅ Sliding Window — Time-based windows
• ✅ Distributed Limiter — Redis-based coordination

Quick Start

from rate_limiter import TokenBucketLimiter

# 100 requests per second, burst of 10
limiter = TokenBucketLimiter(
    rate=100,
    capacity=10
)

# Check if request allowed
if limiter.allow_request():
    process_request()
else:
    return_429_error()

Algorithms

1. Token Bucket

limiter = TokenBucketLimiter(rate=10, capacity=20)

for i in range(25):
    if limiter.allow_request():
        print(f"Request {i}: Allowed")
    else:
        print(f"Request {i}: Rate limited")

2. Sliding Window

limiter = SlidingWindowLimiter(
    window_size=60,  # 60 seconds
    max_requests=100
)

3. Distributed Rate Limiting

from distributed_limiter import RedisRateLimiter

limiter = RedisRateLimiter(
    redis_client=redis_client,
    key="api:user:123",
    max_requests=1000,
    window=60
)

if limiter.allow_request():
    # Process request
    pass

Use Cases

• API throttling
• DDoS protection
• Fair resource allocation
• Cost control

Performance

Algorithm	Memory	Accuracy	Distributed
Token Bucket	O(1)	Exact	Challenging
Leaky Bucket	O(1)	Exact	Challenging
Sliding Window	O(W)	Exact	Yes (Redis)

Standalone Installation

pip install git+https://github.com/navinBRuas/_DistributedSystems#subdirectory=rate-limiting

Usage

See the sections above and examples/ for usage patterns.

Configuration

No mandatory configuration. Optional settings are documented in the package code and examples.

Version

0.1.0 (see VERSION.md)

Changelog

See CHANGELOG.md.

License

MIT License. See repo root LICENSE.