cold cache
A cache that is empty or nearly empty, so almost every request misses and goes to the origin.
In more detail
Caches go cold after deploys, restarts, node replacement or mass eviction, and the origin suddenly sees many times its warm load. Warm the cache before taking traffic and restart in waves.
Example
In the simulation, the first tick sends nearly 300 reads to an origin that sees 15 once warm: a nineteen times spike.Related terms
- miss rate — The share of requests a cache cannot answer, which all go on to the origin.
- origin server — The server that actually holds the content or runs the application, as opposed to a proxy or cache in front of it.
- rolling update — Replacing instances of an application with the new version a few at a time while the rest keep serving.
Taught in
- Latency Numbers Every Engineer Should Know — Scalable System Design & Distributed Architecture
- Caching Strategies And Topology — Scalable System Design & Distributed Architecture
- Load Balancing And The Failure Modes It Introduces — Scalable System Design & Distributed Architecture
- Observability As A Design Input — Scalable System Design & Distributed Architecture
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture