It is en Vogue to subdivide applications into professionally cut microservices and operate them in Kubernetes clusters. Strictly speaking, a Kubernet cluster is a monolithic system consisting of inseparable services. This means that the individual services cannot be operated or switched off independently of each other. This concept is strongly reminiscent of the 90’s with their large application servers. It would be logical to adapt the operating concept to the software architecture: Offer independently operable infrastructure services, such as service registry, deployment distribution, load balancing… In this lecture, we would like to show how the microservice concept of small, closed, specialized modules can also be implemented in infrastructure services without Kubernetes. We want to present use-case-oriented solution sketches.