Problems / Caching, medium
A 90%-hit cache shields Postgres at 2,500 req/s. At 12 s a deploy flushes the cache, and every read hits the database.
Free to open. Runs in your browser, no sign-up.
Success ≥ 99%, p99 ≤ 500 ms through the flush.
2,500 requests a second for 30 seconds: 90% reads, 10% writes.
What each component does in this design, and what calls it. The simulator mimics each one from these settings.
| Component | Its job here | Settings |
|---|---|---|
| Load balancerLoad balancer | Traffic enters here.Calls API.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| APIApp service | Load balancer forwards to it.Calls Redis cache (checked first), Postgres.Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 3 replicas × 200 slots · timeout 1.0 s |
| Redis cacheCache | API checks it first for reads.An in-memory cache like Redis usually answers in well under a millisecond inside one region. | 90% hit rate |
| PostgresSQL database | API reads it when Redis cache misses. API calls it for writes.Postgres runs one process per connection, which is why poolers like PgBouncer sit in front of busy databases. | 2 vCPU · 40 connections |
Try it first. Open one hint at a time if you get stuck.
The database is sized for misses, not for everything.
What happens to the database while the cache warms up?
Give the database headroom for a cold cache: read replicas or more vCPU.