Caching with @Cacheable

Introduction

@Cacheable moves cache-aside boilerplate into declarative annotations — Spring Cache abstraction backed by RedisCacheManager stores method results keyed by SpEL expression. Amazon product services cache getProductByAsin behind @Cacheable("products").

Configure TTL per cache name, disable caching in tests with @Profile, and always plan eviction on @CacheEvict when data mutates. sync=true on @Cacheable prevents stampede on hot keys.

Declarative cache is not magic — wrong key expression caches collisions; null results need unless="#result == null" handling.

Understanding the topic

Key concepts

• @EnableCaching on configuration class.
• RedisCacheManager + CacheConfiguration TTL.
• @Cacheable cacheNames key SpEL.
• @CacheEvict on updates/deletes.
• @CachePut force update cache.
• sync=true — single-flight on miss.

flowchart LR
  App -->|GET key| Redis
  Redis -->|miss| DB
  DB -->|populate| Redis
  Redis --> App

Step-by-step explanation

1. AOP interceptor wraps @Cacheable method.
2. Compute cache key from name + SpEL.
3. Redis GET — hit returns without method call.
4. Miss executes method; SET result with TTL.
5. Evict removes key on mutation.

Syntax reference

Common commands

• unless skips caching null.
• sync=true expensive but stops stampede.
• Custom RedisCacheConfiguration per cache name TTL.

@Cacheable(cacheNames = "products", key = "#sku", unless = "#result == null")
public Product findBySku(String sku) { ... }

@CacheEvict(cacheNames = "products", key = "#sku")
public void updateProduct(String sku, Product p) { ... }

Informative example

Product service with cache and eviction — Spring Boot 3:

@Service
public class ProductService {
    @Cacheable(cacheNames = "catalog", key = "#sku", sync = true)
    public Product getBySku(String sku) {
        return repository.findBySku(sku).orElseThrow();
    }

    @CacheEvict(cacheNames = "catalog", key = "#sku")
    public void invalidate(String sku) {
        repository.save(/* updated */);
    }
}

Register RedisCacheManager bean with Duration.ofMinutes(30) for catalog cache. Enable caching in @Configuration.

Real-world use

Real-world use cases

• Read-heavy repository methods.
• Expensive aggregation computed once per TTL.
• Reference data (countries, tax rates).
• GraphQL DataLoader backend cache.
• API gateway response cache layer.

Best practices

• TTL per cache volatility.
• Evict on every write path.
• sync=true on viral hot keys only.
• Integration test with @CacheEvict clear.
• Monitor hit rate via custom metrics.
• Document SpEL key scheme.

Common mistakes

• Cache without eviction — stale forever until TTL.
• Caching entities with lazy JPA collections — serialization pain.
• Same cache name different key shapes.
• Caching mutable objects returned by reference.

Advanced interview questions

Summary

@Cacheable = declarative cache-aside.