Sensor information is often streamed from connected machines to help monitor equipment status, efficiency, and safety. You can see this data pattern in smart factories, smart buildings, and even in motorsports with connected race cars. The messages inform race teams so they can make better race strategy and maintenance decisions in real time.
Sensor data messages are typically very small—typically around 10 bytes, and they contain fields like sensor ID and sensor reading value (e.g., temperature, tire pressure, or RPM).
The sensor data message rate is extremely high in motorsports. It’s not unusual to see two million messages per second (m/s) because the state of the race car changes so dramatically within each fraction of a second during the race.
Therefore, the proposed partition calculator inputs for connected race car sensor data streams in motorsports, would look as follows.
- The brokers are of similar capability.
- The load on the brokers’ machines is similar.
- The messages don't diverge too much in size.
- The messages are evenly distributed across all partitions.
- The number of brokers makes sense in this context.
- Brokers have similar latencies between producers and consumers.
- The throughput per producer is less than 10MB/s.
- Individual brokers have less than 40k partitions.
- The cluster has less than 200k partitions in total.
