Modeling and Analyzing Architectural Diversity of Open Platforms

Nowadays, successful software companies attain enhanced business objectives by opening their platforms to thousands of third-party providers. When developing an open platform many architectural design decisions have to be made, which are driven from the companies’ business objectives. The set of decisions results in an overwhelming design space of architectural variabilities. Until now, there are no architectural guidelines and tools that explicitly capture design variabilities of open platforms and their relation to business objectives. As a result, systematic knowledge is missing; platform providers have to fall back to ad-hoc decision-making; this bears consequences such as risks of failure and extra costs. In this paper, we present a pattern-driven approach called SecoArc to model diverse design decisions of open platforms and to analyze and compare alternative architectures with respect to business objectives. SecoArc consists of a design process, a modeling language, and an automated architectural analysis technique. It is implemented and ready-to-use in a tool. We evaluate the approach by means of a real-world case study. Results show that the approach improves the decision-making. Future platform providers can reduce risks by making informed decisions at design time.