Caching Strategies for an Erlang Based Web Stack
summary
Two open source caching tools are presented: a memcached client layer that pairs with mcashd + Moxy for distributed, restart-surviving caches, and an in-node Erlang cache er_cache that does per-key eviction, auto-refresh with error-awareness, and macro-based memoization. Deployed on a stack serving 200M monthly uniques, a costly "get related labels" call drops from ~100ms → ~2–4ms, caches ~2k keys with 96.5–98.5% hit ratio, and memory spikes fall (eviction bug fixed from ~11GB). Both solutions are production-tested and available as repos.
AI-written summary · about
chapters (8)
- 0:00Start
- 3:11Storage Layer
- 3:58The Role of Caching
- 6:07Considerations When Choosing a Caching System
- 13:51Airmailed Cash
- 22:51Idea behind this Project
- 27:03Memory Usage
- 33:29Response Times
related talks
32mCache me if you can: Boosted application performance with Redis and client-side caching
49mCache à la carte: a framework for in-memory caching
54mEVCache: Lowering Costs for a Low-Latency Cache with RocksDB
53mLeveraging Caching For Cloud Native Applications
45mWhere is my cache? Architectural patterns for caching microservices by example
13mBMC: Accelerating Memcached using Safe In-kernel Caching and Pre-stack Processing
23mScaling Memcache
16mNyxCache: Flexible and Efficient Multi-tenant Persistent Memory Caching