Cache Tower

Allows stacking cache layers from fast in-memory to persistent Redis or file-based caches, inspired by Stack Overflow's architecture, for optimal performance and resilience across application restarts.

Automatically refreshes stale cache entries in the background to prevent expired-data cache misses, improving throughput without blocking requests, as detailed in the README's example flow.

Redis extensions provide distributed locking and eviction, synchronizing caches across multiple application instances to avoid cache stampedes, essential for scalable microservices.

Supports JSON (Newtonsoft.Json, System.Text.Json) and Protobuf serializers, offering flexibility for different performance and compatibility needs, with specific packages for each.