Masterarbeit von Nektarios Machner

Assessing the cost and benefit of a microservice landscape discovery method in the automative industry

Abstract

Enterprise Architecture Management is a widespread practice in today’s organizations all over the world. Part of this managerial function is to document the current IT landscape in order to provide valuable information, on which decisions are based regarding the future orientation of the enterprise.

Due to the rising adoption of trends like DevOps and microservicebased architectures, Enterprise Architecture Documentation (EAD), which is still performed manually to a large extent, faces various challenges that need to be overcome.

Recent research endeavors explored approaches that assist the documentation process in an automated manner through the usage of runtime data in order to minimize the manual effort and improve the overall quality of available information.

In this thesis, a novel approach to leverage the automated extraction of runtime data from an Application Performance Monitoring (APM) tool and reconstruct the as-is IT landscape is implemented in a real world environment situated in the automotive industry. The implemented IT artifact is tested and evaluated through expert interviews with EA practitioners from our industry partner.

The proposed solution approach shows promising results and is able to automatically extract EA-related information from runtime data and provide various visualizations for exploring the reconstructed IT landscape.

Keywords: Enterprise Architecture Management (EAM), Enterprise Architecture Documentation (EAD), Application Performance Monitoring (APM), distributed tracing, automated discovery algorithm, automated runtime data extraction, automotive industry

Research Questions

• RQ1: Which IT artifacts and their communication relationships can be discovered through runtime data?
• RQ2: What requirements need to be fulfilled for the proposed solution approach to function as intended?
• RQ3: What are benefits and limitations of the proposed solution approach?
  

Sources

1. Farwick, M., Schweda, C. M., Breu, R., Hanschke, I.: A situational method for semi-automated Enterprise Architecture Documentation. pp. 397–426. Softw Syst Model (2016)
2. Hauder, M.,Matthes, F., Roth, S.: Challenges for automated Enterprise Architecture Documentation. In: Trends in Enterprise Architecture Research (TEAR), pp. 21–39. Springer, Berlin, Barcelona, Spain (2012)
3. Holm, H., Buschle, M., Lagerström, R., Ekstedt, M.: Automatic data collection for enterprise architecture models. Softw. Syst. Model. 1–17 (2012)
4. Buschle, M., Ekstedt, M., Grunow, S., Hauder, M., Matthes, F., Roth, S.: Automating Enterprise Architecture Documentation using an enterprise service bus. In: Americas Conference on Insormations Systems (AMCIS) (2012)