Initial implementation by Julien Ponge at Red Hat, Inc.
This demo shows various aspects of Vert.x around the toy scenario of tyres that send pressure updates to some backend.
It especially shows:
- HTTP APIs to store and query data in MongoDB,
- implementation of a edge service with RxJava2 that aggregates data from 3 other HTTP services,
- Kafka-based event streaming with live updates to a web application plugged on the Vert.x event-bus.
Just enjoy live reload
# In one terminal
$ docker compose-up
# In a second terminal
$ ./gradlew vertxRun
# In a third terminal, have fun with the APIs:
$ http POST :3000/ingest tyreId=123 pressure:=12.5
# (...)
The Kafka example is not running from the vertxRun task to avoid polluting the logs, you can just run each instance from the main method of the classes in demo.eventdriven from your IDE.
See locustfile.py with sections to uncomment to generate traffic to the HTTP APIs with Locust.
It's not worth focusing on the reported data, since all instances are running as single verticles, and everything happens on the same machine anyway. I also have no idea of how correct Locust is, especially with respect to coordinated omission.