Backend Caching cookbook

10 recipes · 70 steps · about 51 minutes of reading

Add a cache-aside layer in front of one query
My busiest endpoint hits Postgres for the same query on every single request, and the database is starting to show it.
BeginnerRead 4 minDoing: an hour6 steps
Fix a cache that shows old data
A user updates their profile, refreshes the page, and still sees the version from before the update.
Beginner to intermediateRead 7 minDoing: 30 to 45 minutes6 steps
Namespace your cache keys so two different lookups never collide
Two completely different lookups end up sharing the exact same cache key, and one silently overwrites the other.
Beginner to intermediateRead 5 minDoing: about an hour6 steps
Turn on negative caching so a missing lookup stops hitting Postgres
A bot keeps requesting product IDs that don't exist, and every single one of those requests still reaches your database.
IntermediateRead 5 minDoing: an hour7 steps
Survive a cache stampede under load
The moment your most popular cache key expires, hundreds of requests pile onto the database in the exact same second.
IntermediateRead 6 minDoing: an afternoon8 steps
Choose local caching or Redis
Everything works with one server and a plain in-memory dictionary as a cache, then you add a second server and the two stop agreeing.
IntermediateRead 4 minDoing: half an hour6 steps
Move a user's session lookup out of Postgres
Every single authenticated request runs a database query just to check who's making it.
IntermediateRead 4 minDoing: an afternoon6 steps
A Redis eviction policy that silently drops the keys your app still needs
Redis is either silently forgetting keys you still need, or it has stopped accepting writes altogether, and you don't know which policy is even set.
Intermediate to advancedRead 5 minDoing: an afternoon8 steps
Build a sliding window rate limiter for one endpoint
Nothing stops one client from hammering your busiest endpoint hundreds of times a second.
Intermediate to advancedRead 5 minDoing: an afternoon8 steps
Prove your cache is faster, never slower
You added Redis because that is what you are supposed to do, and you genuinely don't know if the endpoint is any faster.
AdvancedRead 6 minDoing: a day9 steps

Cookbook badge

Finish all 10 recipes to earn this.

In these roadmaps

Backend Developerstep 9 of 15

Backend Caching cookbook

Add a cache-aside layer in front of one query

Fix a cache that shows old data

Namespace your cache keys so two different lookups never collide

Turn on negative caching so a missing lookup stops hitting Postgres

Survive a cache stampede under load

Choose local caching or Redis

Move a user's session lookup out of Postgres

A Redis eviction policy that silently drops the keys your app still needs

Build a sliding window rate limiter for one endpoint

Prove your cache is faster, never slower

In these roadmaps

Backend Caching cookbook

Add a cache-aside layer in front of one query

Fix a cache that shows old data

Namespace your cache keys so two different lookups never collide

Turn on negative caching so a missing lookup stops hitting Postgres

Survive a cache stampede under load

Choose local caching or Redis

Move a user's session lookup out of Postgres

A Redis eviction policy that silently drops the keys your app still needs

Build a sliding window rate limiter for one endpoint

Prove your cache is faster, never slower

In these roadmaps