Problems / Async and streaming, hard
The topic has 2 partitions and the consumer group processes each event in about 5 ms. Events arrive at 2,000 a second and are kept for 30 seconds.
Free to open. Runs in your browser, no sign-up.
No data lost to retention, success ≥ 99.9%.
2,000 requests a second for 45 seconds: 100% event (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 Ingest API.Layer 4 balancers route by IP and port; layer 7 balancers read HTTP and can route by path or header. | – |
| Ingest APIApp service | Load balancer forwards to it.Calls Event stream (publishes).Little's law: throughput ≈ concurrency ÷ latency. 200 slots at 50 ms serve about 4,000 req/s. | 2 replicas × 200 slots · timeout 1.0 s |
| Event streamEvent stream | Ingest API publishes to it.Calls Consumer group.Kafka keeps order only within a partition, so the partition key decides what stays in sequence. | 2 partitions · retention 30 s |
| Consumer groupWorker | Event stream consumes from it.Queues deliver at least once, so a worker can see the same message twice. | 2 concurrent jobs |
Try it first. Open one hint at a time if you get stuck.
Parallelism is capped by partitions: two consumers at most.
2 consumers ÷ 5 ms = 400 events a second; you need 2,000.
More partitions, then enough consumers to match.
Add partitions ("12 partitions"), then consumers up to the partition count.