Mahboob-A / rate-limiter

Rate Limit PoC implemented using Token Bucket Algorith,

Rate Limiter

This is an pythonic example of TokenBucket algorithm to implement rate limiter.

General Information

A rate limiter is a crucial component in software application when the resources are exposed to outside world and the stakeholders want to control how the resources are being manipulated.

A rate limiter can be implemented to prevent any DoS or DDoS attack from malicious parties. Dos attack is continious demand of computing resources making evenly resouce distribution impossible and room for resource starvation.

A rate limiter is also often implemented when the stakeholders want to limit the number of requests permitted to evenly distribute the computing resources to all the parties.

Documentation

Input Style - Terminal

The bucket size i.e. the total tokens in the bucket.

token_size 

Frequency after how many minutes the bucket will be auto reset.

refill_rate_time_in_minutes

How many seconds to wait for the next request to the api.

seconds_to_wait_to_simulate_request

Code Structure

The main interface to the resource - a calculator

api.py 

The logic to limit the request frequency to the api to access the calculator

rate_limiter.py

The resource to share through the api - functionalities of a calculator

calculator.py

Run Locally

Clone the project

  https://github.com/Mahboob-A/rate-limiter.git

Go to the project directory

  cd rate_limiter

Start the server

• for Linux/MacOS

    python3 api.py 

• for windows

    python api.py 

Optimizations

The tokenbucket algorithm implemented here is slightly different than the standard how it is done.

I have customized it while analyzing the requirements.

FAQ

Which algorithm has been implemented in the rate limier

TokenBucket Algorithm

Language

Python

Screenshots

API startup and Acceptance of requests

Request throttled and Bucket reset after refil time