The proxy design pattern allows scalability using both the x-axis and the z-axis, and everything is referenced by how many instances of a service are available for proxy access.
According to the configuration of our Nginx, which makes the proxy role for us, we've four instances of each microservice. This can be verified by the number of references to a microservice instance that is in the configuration of our upstream, as shown in the following example:
upstream users_servers { server bookproject_usersservice_1:3000; server bookproject_usersservice_2:3000; server bookproject_usersservice_3:3000; server bookproject_usersservice_4:3000; } upstream orcherstrator_servers { server bookproject_orcherstrator_news_service_1:5000; ...
