I built Polar Bear Cache, a local caching library for Java and Spring applications, with a simple idea: serve frequently accessed data locally and reduce repeated trips to an external cache.
It integrates with Spring’s familiar @Cacheable, @CachePut, and @CacheEvict annotations. Each service instance keeps its own local cache. With an event service configured, instances broadcast invalidation notifications instead of cached values, allowing other instances to discard affected entries.
Local cache hits avoid network round trips and repeated object deserialization. For suitable workloads, that can reduce pressure on external cache infrastructure and help each service instance handle more concurrent requests.
The best fit is data that is read frequently, changes less often, and can tolerate short delays in cache invalidation. Cross-instance invalidation is asynchronous, so it does not guarantee strong consistency.
So, why “Polar Bear Cache”?
The name comes from a slightly optimistic chain of engineering reasoning:
More local cache hits
→ Fewer remote cache requests
→ More traffic handled per service instance
→ Potentially fewer servers
→ Less energy consumption and fewer greenhouse gas emissions
→ Help save polar bears! 🐻❄️🧊
Of course, caching alone won’t save the Arctic. Actual savings depend on the workload and infrastructure. But making software more resource-efficient feels like a worthwhile place to start.
That’s the idea behind Polar Bear Cache.
Curious about how it works? Check out Polar Bear Cache on GitHub.
Feedback, ideas, and contributions are welcome!
Fewer remote calls. More ice.
Top comments (0)