Problems / LLM serving, medium
A support-agent fleet sends 30 requests a second through a prefix-aware router to 6 model replicas. 85% of requests start with the same 12,000-token system prompt and tool schemas, and the router always sends a prefix to the replica that holds it.
Unlock it once for 15 credits; everything inside it is free (runs, changes, Submit, results, hints). Sign in with Google to start; new accounts get free credits.
Success ≥ 99%, p99 ≤ 3 s.
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 6; 2 of them are hidden until then.
30 requests a second for 30 seconds: 100% request (reads).
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 AI app.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| AI appApp service | Load balancer forwards to it.Calls Prefix-aware router.Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 3 replicas × 500 slots · timeout 30 s |
| Prefix-aware routerInference router (KV-aware) | AI app calls it.Calls GPU replica 1, GPU replica 2, GPU replica 3, GPU replica 4, GPU replica 5, GPU replica 6.Routing a session back to the replica that holds its prefix can raise KV-cache hit rates from about 20% to 75%. | 100000 slots |
| GPU replica 1Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |
| GPU replica 2Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |
| GPU replica 3Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |
| GPU replica 4Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |
| GPU replica 5Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |
| GPU replica 6Self-hosted serving pool | Prefix-aware router forwards to it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 4 slots |