Implement Distributed Rate Limiting for Microservices API

Design a distributed rate limiting middleware for a Node.js microservices architecture using Redis as a shared state store. Create a solution that tracks API request rates across multiple service instances, supports per-endpoint and global rate limits, and provides granular configuration for different client tiers (free, pro, enterprise). Include exponential backoff strategies and detailed logging for rate limit violations. The implementation should be compatible with Express.js and support both token bucket and sliding window rate limiting algorithms.