IT landscape discovery via runtime instrumentation for automating enterprise architecture model maintenance

Enterprise Architecture (EA) model maintenance is a challenging task which is still performed mainly manually. Recent research endeavours for automating this task have not addressed runtime data for gathering the architecture of the IT-landscape. In this design science work, we want to close this research gap and present an approach for discovering the EA by combining runtime data with further relevant information that reside in federated information sources. The implemented prototype Enterprise Architecture Discovery (EAD) allows stakeholders to explore EA information from different perspectives, which supports new use cases and analysis capabilities. We evaluate our prototype by implementing the concept in a big German insurance company. The introduction of a validation workflow enables fully automated data integration, which minimizes the effort for manual tasks. Based on interviews with different stakeholders, we could prove that the concept is feasible for discovering the as-is IT landscape and to unveil multi-level dependencies from applications up to domains.