Caching Strategy Implementation Guide

You are a performance engineer who implements caching systems. Design a caching strategy for this application.

Application: [DESCRIBE APPLICATION]
Current bottlenecks: [WHAT'S SLOW]
Data characteristics:
- Read/write ratio: [RATIO]
- Data freshness needs: [REAL-TIME / MINUTES / HOURS]
- Data size: [SIZE]

Infrastructure: [REDIS / MEMCACHED / CDN / IN-MEMORY]
Scale: [CONCURRENT USERS]

Design:

**Caching Layers**
```
Browser Cache → CDN → Application Cache → Database Cache → Database
```

**Layer 1: Browser/Client Cache**
- Cache-Control headers
- ETags implementation
- Service worker caching

**Layer 2: CDN Caching**
- What to cache at edge
- Cache key design
- Purge strategy

**Layer 3: Application Cache**
```[language]
// Redis/Memcached implementation
const cache = {
  async get(key) { },
  async set(key, value, ttl) { },
  async invalidate(pattern) { }
};
```

**Cache Key Design**
```
{namespace}:{entity}:{id}:{variant}
Example: users:profile:123:full
```

**TTL Strategy**
| Data Type | TTL | Rationale |
|-----------|-----|----------|
| User profile | 5 min | Changes rarely |
| Product list | 1 min | Updates frequently |

**Cache Invalidation**

1. Time-based expiration
2. Event-based invalidation
```[language]
// Publish invalidation event
// Subscribe and clear cache
```

3. Write-through/write-behind patterns

**Cache Patterns**

- Cache-aside (lazy loading)
- Read-through
- Write-through
- Write-behind

```[language]
// Implementation for each pattern
```

**Stampede Prevention**
- Locking mechanism
- Probabilistic early expiration
- Background refresh

**Monitoring**
- Hit/miss ratios
- Memory usage
- Eviction rates
- Latency percentiles

**Failure Handling**
- Cache unavailability
- Graceful degradation
- Circuit breaker for cache

You are a performance engineer who implements caching systems. Design a caching strategy for this application.

Application: [DESCRIBE APPLICATION]
Current bottlenecks: [WHAT'S SLOW]
Data characteristics:
- Read/write ratio: [RATIO]
- Data freshness needs: [REAL-TIME / MINUTES / HOURS]
- Data size: [SIZE]

Infrastructure: [REDIS / MEMCACHED / CDN / IN-MEMORY]
Scale: [CONCURRENT USERS]

Design:

**Caching Layers**
```
Browser Cache → CDN → Application Cache → Database Cache → Database
```

**Layer 1: Browser/Client Cache**
- Cache-Control headers
- ETags implementation
- Service worker caching

**Layer 2: CDN Caching**
- What to cache at edge
- Cache key design
- Purge strategy

**Layer 3: Application Cache**
```[language]
// Redis/Memcached implementation
const cache = {
  async get(key) { },
  async set(key, value, ttl) { },
  async invalidate(pattern) { }
};
```

**Cache Key Design**
```
{namespace}:{entity}:{id}:{variant}
Example: users:profile:123:full
```

**TTL Strategy**
| Data Type | TTL | Rationale |
|-----------|-----|----------|
| User profile | 5 min | Changes rarely |
| Product list | 1 min | Updates frequently |

**Cache Invalidation**

1. Time-based expiration
2. Event-based invalidation
```[language]
// Publish invalidation event
// Subscribe and clear cache
```

3. Write-through/write-behind patterns

**Cache Patterns**

- Cache-aside (lazy loading)
- Read-through
- Write-through
- Write-behind

```[language]
// Implementation for each pattern
```

**Stampede Prevention**
- Locking mechanism
- Probabilistic early expiration
- Background refresh

**Monitoring**
- Hit/miss ratios
- Memory usage
- Eviction rates
- Latency percentiles

**Failure Handling**
- Cache unavailability
- Graceful degradation
- Circuit breaker for cache