Writing from the build
This is where we think out loud about the decisions behind Aeron Cache: why we built a cache on top of low-latency messaging infrastructure, what mechanical sympathy really buys you in a data store, and how streaming deltas instead of invalidations changes the systems you can design. We write for the engineers who choose infrastructure and have to live with it afterward, so expect strong opinions stated plainly and kept technically honest.
Articles
- Why we built Aeron Cache — The origin story: the gap between the messaging stack we trusted for microseconds and the caches we settled for everywhere else, and the bet that closing it was worth it.
- The case for mechanical sympathy in caching — Allocation discipline, data-oriented layout, and single-threaded replicated state machines, and why they are the difference between a predictable tail and a hopeful one.
- Patch-native streaming: caches that speak deltas — What changes when your cache streams only the fields that changed instead of shouting “this key is stale” — with concrete JSON before and after.