Can you recommend a cache store for rate limiting in a high-throughput API?

Yes — for a high-throughput API, the best cache store for rate limiting is usually Redis.

Why Redis is a strong choice

• Very fast: in-memory reads/writes with low latency
• Atomic operations: supports commands and Lua scripts for safe increment/check logic
• TTL support: easy to expire counters automatically
• Widely used: lots of client libraries and battle-tested patterns
• Scales well: can be deployed as a cluster for larger workloads

Good alternatives

• Memcached: fast and simple, but less ideal because it lacks rich atomic/TTL patterns compared to Redis
• KeyDB: Redis-compatible, often used for higher throughput in some setups
• Aerospike: good for very large-scale distributed caching, but more operational complexity
• In-process memory: only suitable for single-instance or local throttling, not distributed APIs

Recommendation by use case

• Most APIs: Redis
• Very high throughput / multi-region complexity: Redis Cluster or KeyDB
• Extremely large scale with specialized infra: Aerospike

Tip

If you use Redis, implement rate limiting with:

• INCR + EXPIRE for simple fixed-window limits
• Lua scripts for atomic logic
• SET with expiration for token-bucket or sliding-window approaches

If you want, I can also suggest:

1. the best rate-limiting algorithm, or
2. a Redis schema / implementation example.