I have a requirement to use Kafka for SSE events, my application is a mid-tier application which hooks up Mobile chat calls from front-end with the server-side.
P.S: I saw a post similar to my requirement "https://stackoverflow.com/questions/65071532/replay-kafka-topic-with-server-sent-events" I have a few questions on it.
If we go with Kafka, is it a good practice to use multiple group id's, if an application is deployed in 200 machines(creating group id with machine name), as it will create 200 group id's?
Whenever we get a response from the server for a particular call we will need to map it to the same machine, in order to achieve it we have to consume messages in all 200 machines and read the header to see if it belongs to the same machine if machine name matches then we will send the response but other machines which consumes the same have to discard it after reading the header, will it not be an over kill in terms of CPU consumption or memory usage?