Problems / Real world, hard
Asked at Meta (WhatsApp, Messenger) and Slack. Every message is written to Postgres and then pushed to the recipient’s phone in the same request; the app polls the same database to sync, and presence heartbeats are database writes too. At 12 s the push provider is down for 8 seconds, and sending stops.
Unlock it once for 25 credits; everything inside it is free (runs, changes, Submit, results, hints). Sign in with Google to start; new accounts get free credits.
Success ≥ 99.5%, p99 ≤ 600 ms and Apdex ≥ 0.90 through the push outage, for at most $46,000 a month.
Like a coding problem, a solution has to pass every case, not just the one it was built for. Submit in the lab runs all 3; 1 of them are hidden until then.
20,000 requests a second for 20 seconds: 50% send (writes), 40% sync (reads), 10% presence (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 Chat API.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| Chat APIApp service | Load balancer forwards to it.Calls Messages (Postgres), Mobile push (APNs, FCM).Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 12 replicas × 100 slots · timeout 2.0 s |
| Messages (Postgres)SQL database | Chat API calls it.Postgres runs one process per connection, which is why poolers like PgBouncer sit in front of busy databases. | 16 vCPU · 800 connections |
| Mobile push (APNs, FCM)Notification service | Chat API calls it for /send.Email, SMS and push providers all throttle senders and can delay delivery. | 200 slots · 2,000,000 calls/min |