Real-time APIs

We have introduced some web API styles, such as REST-based APIs, gRPC APIs, and GraphQL APIs. They all follow the request/response pattern – the client sends a request to the server, and the server returns a response. This pattern is easy to understand. However, this pattern may not be suitable for some scenarios.

Let’s say we have an application that contains two parts – the server, which is a fire station, and the clients, which are the fire engines. How can we notify the fire engines when an incident occurs?

If we use the request/response pattern, the clients need to send requests to the server to get the latest notification regarding the incident. But what is the best frequency of sending requests? 1 minute, or 10 seconds? Think about how urgent the incident is. If the fire engine gets a notification after a 10-second delay, this might be a problem, as the fire might be more serious and urgent! So, what about sending requests every 1 second...