As we do not use the Publish-Subject approach and Spring's @EventListener annotation, we do not depend on Async Support, so the application configuration becomes simpler:

@SpringBootApplication
public class Application {
   public static void main(String[] args) {
      SpringApplication.run(Application.class, args);
   }
}

As we can see, this time we do not need to enable Async Support using the @EnableAsync annotation, and we also do not need to configure Spring's Executor for event handling. Of course, if required, we may configure a RxJava Scheduler for fine-grained thread management when processing reactive streams, but such a configuration would not depend on Spring Framework.

In turn, we do not need to change the code for the UI part of the application; it should work the same way as before. Here, we have to highlight that with the RxJava-based implementation, the temperature sensor is not probed when nobody listens. Such behavior is a natural consequence of the fact that reactive programming has a notion of active subscription. The Publish-Subject-based implementations do not have such properties, and as such are more limited.