Design a rate limiter.

Focus on the core functionality of a rate limiter, which is to control the number of requests a user can make to a service within a specified time frame. Discuss different algorithms for implementing rate limiting, such as Token Bucket or Leaky Bucket, and explain the trade-offs between them. Consider scalability and performance, especially in high-traffic scenarios, and how you would handle edge cases like burst traffic. Additionally, think about how to store state (in-memory vs. distributed) and how to ensure fairness among users. Be prepared to address potential challenges, such as handling retries and ensuring the system is resilient to abuse.