Problems / AI products, hard
Legal questions traverse a citation graph: 12 ms of CPU each on 4 vCPUs, at 500 questions a second.
Free to open. Runs in your browser, no sign-up.
Success ≥ 99%, p99 ≤ 5 s.
500 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 Citation graph (GraphRAG), GPU serving pool.Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 3 replicas × 500 slots · timeout 30 s |
| Citation graph (GraphRAG)GraphRAG index | AI app calls it.GraphRAG builds a knowledge graph from documents so multi-hop questions can follow relationships. | 64 slots |
| GPU serving poolSelf-hosted serving pool | AI app calls it.GPU servers batch requests together; throughput rises with batch size until memory runs out. | 2000 slots |
Try it first. Open one hint at a time if you get stuck.
500 × 12 ms = 6 vCPUs.
Find the red card first, then its capacity: slots ÷ seconds per request.
Give the GraphRAG index more capacity.