Spring Boot Cache Example

Implementing caching via Redis in a Spring Boot + MySQL application can massively boost performance and reduce DB load, but it requires a careful design to avoid stale data, inconsistency, and cache stampede.

---

✅ 1. Design Considerations

🔧 General Design

• Read-heavy queries → cacheable in Redis

• Cache key-value pairs, typically:

  • Key: user:123 or product:sku:abc
  • Value: JSON-serialized object (or Hash/String)

🧱 Common Caching Patterns

Pattern	Description
Read-Through	App checks cache first → DB fallback on miss → populate cache
Write-Through	Writes to cache and DB at the same time
Write-Behind (Async)	Writes to cache, flushes to DB later (riskier)
Cache-Aside (Lazy Load)	App reads cache, falls back to DB, populates cache if missed (most common)

✅ Recommended: Cache-Aside (Spring Boot default pattern)

@Cacheable(value = "user", key = "#userId")
public User getUserById(Long userId) {
    return userRepository.findById(userId).orElseThrow();
}

---

✅ 2. Ensuring Consistency (DB <-> Cache)

The key challenge: When DB changes, how do we keep Redis cache in sync?

🔁 Sync Strategies

Strategy	How It Works	Pros	Cons
Evict on Write (@CacheEvict)	After DB write, remove the cache entry	Simple, consistent	Requires cache reload on next read
Update Cache on Write (@CachePut)	Write to DB and cache together	Fast reads	Risk of partial failure (DB updated but cache not)
Message Queue (e.g. Kafka)	Async sync via event stream	Decoupled, scalable	Complexity, potential lag
Double Write Transaction	Transactionally update both	Reliable if done right	Harder to manage rollback

🔐 Spring Boot Code Example: Cache Eviction on Update

@CacheEvict(value = "user", key = "#user.id")
public void updateUser(User user) {
    userRepository.save(user); // DB write
}

---

⚠️ 3. Bottlenecks & Edge Cases

💥 Cache Stampede

Many requests hit the DB when cache expires

• ✅ Solution: Use randomized TTL + mutex locks (SETNX)
• Libraries: Caffeine (local), Redisson

🔁 Cache Inconsistency

DB updated, cache not updated

• Use @CacheEvict, @CachePut, or transactional messaging
• Avoid parallel updates — lock or debounce as needed

🧊 Cold Start

Cache empty after restart / deployment

• ✅ Pre-warm cache during boot (optional batch loaders)

🔓 Serialization Format

• Redis needs fast serialization
• Prefer JSON (e.g. Jackson2JsonRedisSerializer) or String over Java native serialization

@Bean
public RedisTemplate<String, Object> redisTemplate(RedisConnectionFactory factory) {
    RedisTemplate<String, Object> template = new RedisTemplate<>();
    template.setConnectionFactory(factory);
    template.setValueSerializer(new Jackson2JsonRedisSerializer<>(Object.class));
    return template;
}

---

🧪 Example Caching Flow in Spring Boot (Cache-Aside):

@GetMapping("/user/{id}")
@Cacheable(value = "user", key = "#id")
public User getUser(@PathVariable Long id) {
    return userService.findById(id);  // Calls DB only if cache miss
}

@PutMapping("/user")
@CacheEvict(value = "user", key = "#user.id")
public void updateUser(@RequestBody User user) {
    userService.update(user); // Removes stale cache
}

---

🧠 Summary Checklist

Goal	Technique
Improve read performance	Cache-Aside w/ Redis
Keep cache & DB in sync	Use @CacheEvict on write
Avoid stampede / cold start	Pre-warm or use lock/mutex
Handle failures	Use retry logic, fallbacks
Choose right TTL	Add random jitter to avoid stampede
Fast serialization	Use JSON or String serializers