Microservices are all about breaking things down to individual components. Having so many components results in many repositories and thus into many deployment cycles. Also, we are talking about speed, which increases the number of frequent deployment on a daily basis. As a ripple effect of this, infrastructure operation work may increase such as they have to maintain the deployment speed, to create separate deployment pipeline for each and every microservice, making sure that one microservice is not affecting another microservice during deployment and so on. If you have services more than 20 that talk to a new microservice, testing all these microservices would take of a lot of time. This concludes to us the solution that we should have automatic testing, which, in turn, starts impacting the resources. For example, then, there should be one another environment for testing. More servers means more money. Running all these microservices also uses quite a large number of servers than a monolithic application. As the number of environment/servers increases in an organization, DevOps dependency or involvement also starts increasing. So, high DevOps costs also a common issue here.