Where Is My Cache? Architectural Patterns for Caching ...
summary
Compares five concrete caching patterns and when each actually works: embedded cache (Guava/Ehcache with Spring @Cacheable) for lowest JVM latency but JVM‑bound; embedded distributed Hazelcast to share cache across instances with discovery plugins; client‑server (Redis/memcached or Hazelcast client) for language‑agnostic ops at higher network latency; sidecar per pod in Kubernetes for localhost performance plus isolation; and reverse‑proxy/NGINX or reverse‑proxy sidecar for declarative, code‑free caching. Highlights tradeoffs (millisecond sensitivity, separate ops, cache invalidation) and gives a simple decision tree plus setup tips and a 300‑byte vs multi‑MB update performance question.
AI-written summary · about
chapters (15)
- 0:00Introduction to caching patterns
- 2:25Why caching in microservices
- 4:53Embedded cache pattern
- 9:55Embedded distributed cache demo
- 13:36Pros and cons of embedded cache
- 13:57Client-server cache pattern
- 17:14Cloud cache as a service
- 20:58Client-server and cloud cache pros/cons
- 23:35Sidecar cache pattern in Kubernetes
- 27:12Sidecar cache pros and cons
- 28:21Reverse proxy cache pattern
- 31:17Reverse proxy sidecar cache demo
- 36:10Reverse proxy sidecar limitations
- 37:01Quick decision tree for caching
- 39:19Q&A on cache updates
related talks
25mArchitectural Caching Patterns for Kubernetes
46mA Hitchhiker’s Guide to Caching Patterns
43mCache Me Outside - Caching Methodologies & Architectures
45mCaching in Applications Still Matters
50mLet's talk about the cache!
46mCaching Made Bootiful
44mCatch Me If You Cache: Cache Patterns and Pitfalls in Production
01h 10mCaching with Spring: Advanced Topics and Best Practices