Understanding Caching in Spring: Local vs Distributed Cache and GuavaCacheManager

Cache is a memory‑based storage used to improve performance by avoiding frequent database reads, which involve slow disk access.

Two main cache types are local cache (stored in the same JVM, e.g., static variables or a locally deployed Redis) and centralized cache (e.g., Redis cluster) that provides high availability and consistency across multiple servers.

Local cache offers low latency but low reliability; centralized cache adds network latency but ensures data persistence and consistency.

Spring provides cache abstraction via CacheManager; commonly used implementations include ConcurrentMapCacheManager, EhCacheCacheManager, GuavaCacheManager, and RedisCacheManager.

GuavaCacheManager uses LRU + ConcurrentHashMap and can be configured in Spring Boot. Example Maven dependencies:

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-cache</artifactId>
</dependency>
<dependency>
    <groupId>com.google.guava</groupId>
    <artifactId>guava</artifactId>
    <version>18.0</version>
</dependency>

Configuration class:

@EnableCaching
@Configuration
public class GuavaCacheConfig {
    @Bean
    public CacheManager cacheManager() {
        GuavaCacheManager cacheManager = new GuavaCacheManager();
        cacheManager.setCacheBuilder(
            CacheBuilder.newBuilder().expireAfterWrite(3, TimeUnit.MINUTES));
        return cacheManager;
    }
}

Cache expiration strategies include expireAfterAccess, expireAfterWrite, and refreshAfterWrite.

The @Cacheable annotation can be used to cache method results; its key and value are defined by EL expressions, and the cached value is the method’s return value.

Cache miss may occur when a cached method is called from within the same class because Spring’s AOP proxy is bypassed.

In summary, combining local and centralized caches, configuring GuavaCacheManager, and using Spring’s caching annotations can improve application performance while being aware of pitfalls such as internal method calls that bypass the proxy.