cache_patterns

Add the spring-boot-starter-cache dependency, annotate a configuration class with @EnableCaching, and declare a CacheManager bean such as CaffeineCacheManager. Then annotate service methods with @Cacheable to store and reuse results.

@Cacheable reads from the cache and only executes the method on a miss. @CachePut always executes the method and refreshes the cached value. @CacheEvict removes entries, optionally clearing all entries with allEntries=true.

Caffeine suits single-instance, in-memory caching with low latency. Redis fits distributed systems where multiple service instances must share cache state. Both integrate through Spring's CacheManager abstraction without changing business code.

No, you should not mix Spring annotations like @Cacheable with JCache annotations like @CacheResult on the same method. Choose one annotation model per method to avoid conflicting interception behavior.

Enable the Actuator cache endpoint to inspect cache statistics, and push metrics through Micrometer for production monitoring. Keep hit/miss logging at debug level to avoid log noise.

Stale data occurs when eviction is not aligned with transactional boundaries or when @CachePut is missing on update methods. Align cache eviction with transaction commits and use publish/subscribe invalidation for multi-instance deployments.