BEAR.QueryRepository is a distributed caching framework for BEAR.Resource applications, inspired by CQRS. It segregates reads and writes into separate repositories to optimize performance and resource utilization.
- Event-Driven Cache Invalidation: Automatically invalidates cache when data changes, ensuring consistency.
- Dependency Resolution: Resolves dependencies between resources and updates related caches automatically.
- Donut Caching: Combines dynamic and static content for efficient partial caching.
- CDN Integration: Seamlessly integrates with modern CDNs (e.g., Fastly, Akamai) for shared cache management.
- Conditional Requests with ETag Support: Reduces network overhead by leveraging
ETagand304 Not Modifiedresponses. - Distributed Cache Support: Works with server-side caches (e.g., Redis, APC), shared caches (e.g., CDNs), and client-side caches.
- BEAR.Sunday cache manual
- LLM Documentation
- Why the QueryRepository Log Records Everything — design rationale, the cost measurements, and why recording is off by default
- Reading the Log — every word the log can contain, and how to read a session
- What the Cache Log Proves — and What It Does Not — the questions the log answers, the mechanisms that keep the answers honest, and the declared boundaries