Problems / Scaling, medium
Every upload is resized synchronously by an image service: 40 ms of CPU each on 4 vCPUs, at 300 uploads a second.
Free to open. Runs in your browser, no sign-up.
Success ≥ 99%, p99 ≤ 1 s.
300 requests a second for 30 seconds: 0% reads, 100% 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 Image resizer, 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 |
| PostgresSQL database | API calls it.Postgres runs one process per connection, which is why poolers like PgBouncer sit in front of busy databases. | 8 vCPU · 60 connections |
| Image resizerImage processing | API calls it.Resizing images is CPU-heavy, so it is usually done once and cached. | 40 slots |
Try it first. Open one hint at a time if you get stuck.
CPU needed = 300 × 40 ms = 12 vCPUs.
Image work is CPU-bound.
Give the image service more capacity and vCPU (or move resizing to a queue and workers).