caching

Implement Redis caching for NestJS repository adapters by wrapping existing ports with a Redis-backed cached adapter. This introduces a layered caching strategy that preserves data correctness while reducing repeated reads and expensive recomputation across your infrastructure.

Distributed cache invalidation uses Redis client tracking to monitor data mutations across services. When a write operation occurs, the system executes staged invalidation to automatically remove or update stale cache entries, ensuring data consistency without manual cleanup.

Configure frontend caching alongside a Redis backend by utilizing HTTP Cache-Control headers and integrating TanStack Query. This combined approach enables selective client-side caching strategies that reduce redundant network requests while maintaining data freshness.

Yes, a port and adapter boundary is required to set up this caching layer. The architecture explicitly wraps existing repository ports, API clients, or HTTP endpoints with Redis-backed cache stores to intercept reads and apply mutation-invalidation on writes.

Yes, you can apply HTTP Cache-Control headers to external API clients. The caching strategy selectively utilizes these headers on HTTP endpoints and external API clients to enable robust client-side caching and reduce repeated network traffic.

Use a layered caching strategy instead of a single cache to independently optimize read performance across infrastructure adapters, HTTP endpoints, and the frontend. This distributed approach preserves data correctness through mutation-invalidation while preventing broad cache wipes.