kienbui1995/caching-strategy

Caching Strategy

Overview

Caching trades freshness for speed. The hard part isn't adding a cache — it's invalidating it correctly.

Core principle: Cache at the right layer, with the right TTL, and a clear invalidation strategy.

When to Use

• Slow API responses that serve repeated data
• High database load from identical queries
• Static assets delivery optimization
• Reducing third-party API calls

Cache Layers (top to bottom)

| Layer | TTL | Best For | |-------|-----|----------| | CDN (Cloudflare, CloudFront) | Hours-days | Static assets, public pages | | HTTP cache (browser) | Minutes-hours | API responses, images | | Application cache (Redis/Memcached) | Seconds-minutes | DB query results, computed data | | In-process (memoization) | Request lifetime | Expensive computations within a request |

Cache-Aside Pattern (most common)

1. Check cache for key
2. Cache hit → return cached value
3. Cache miss → fetch from source, store in cache, return

Invalidation Strategies

| Strategy | How | Use When | |----------|-----|----------| | TTL expiry | Set expiration time | Data can be slightly stale | | Write-through | Update cache on every write | Strong consistency needed | | Event-driven | Invalidate on data change event | Microservices, async systems | | Cache tags | Tag entries, purge by tag | Related data changes together |

HTTP Cache Headers

# Immutable static assets (hashed filenames)
Cache-Control: public, max-age=31536000, immutable

# API responses (revalidate)
Cache-Control: private, max-age=0, must-revalidate
ETag: "abc123"

# No caching
Cache-Control: no-store

Anti-Patterns

| Pattern | Problem | |---------|---------| | Cache everything | Stale data bugs, memory waste | | No TTL | Cache grows forever | | Cache without invalidation plan | Stale data served indefinitely | | Caching user-specific data in shared cache | Data leaks between users | | Cache stampede | All caches expire simultaneously → DB overload |

Redis Quick Reference

SET user:123 '{"name":"..."}' EX 300    # 5 min TTL
GET user:123
DEL user:123                              # Invalidate

Integration

• magic-powers:performance-optimization — identify what to cache via profiling
• magic-powers:infrastructure-review — review cache infrastructure
• magic-powers:database-optimization — reduce DB load with query caching