An aspect of constructing a network of microservices is the ability to assess the effect that a microservice change has not only with respect to the process for which the change was made, but also with respect to each other process that forms part of the network. In cases in which the same microservice is used by different processes—ranging from sophisticated banking applications to single devices that are part of the Internet of Things—the impact of a change in the microservice can affect each of the other processes that also use the microservice. Viewing microservices as a portfolio of services rather than limiting the view to a single process allows for an assessment of the effect of a microservice change on the portfolio of processes and not merely the effect on a process in isolation.

Another aspect of the inventive arrangements disclosed herein is the combination and compact representation of various classes of data for monitoring microservice changes. The combination and compact representation of data enables a computer system to more efficiently monitor the changes, thus improving the performance efficiency of microservices monitoring systems.

Another aspect of the inventive arrangements disclosed herein is the generation of various analytics for assessing changes in the dynamic behavior of a network of microservices. The analytics can assess the likely impact of microservice modifications and changes to one or more microservices. The analytics can assess the likely source of system errors and subpar performance of processes owing to recent changes in or modifications to one or more microservices. The analytics can enable a DevOps team to identify a microservice that should be rolled back (to a prior version after a new version is introduced) in order to bring a network of microservices back to a “normal” or “well-functioning” state.