Cache is the most effective way to speed up an application. But poor cache invalidation is one of the hardest problems in CS.

Strategies¶

• Cache-aside: Application reads/writes cache explicitly
• Write-through: Write to cache and DB simultaneously
• Write-behind: Write to cache, async to DB
• Read-through: Cache automatically loads from DB

HTTP Cache Headers¶

Static assets — long cache¶

Cache-Control: public, max-age=31536000, immutable

API response — short cache¶

Cache-Control: private, max-age=60

Validation with ETag¶

ETag: “abc123” If-None-Match: “abc123” # 304 Not Modified

Cache-Aside Pattern¶

async def get_product(product_id):

1. Check cache¶

cached = await redis.get(f”product:{product_id}”) if cached: return json.loads(cached)

2. Cache miss — fetch from DB¶

product = await db.products.find(product_id)

3. Store in cache¶

await redis.setex(f”product:{product_id}”, 3600, json.dumps(product)) return product

Key Takeaway¶

Cache-aside for most use cases. HTTP cache headers for CDN. Invalidation is harder than it seems — TTL is the simplest approach.