Growth- and Entropy-Based SOA Measurement: Vision and Approach in a Large Scale Environment

Service-oriented architectures (SOA) can be considered as an enabling methodology for Cloud Computing and the IT's industrialization, standardization and commodity approach. The efforts to effectively introduce SOA were driven by the evermore increasing complexity and volume of information to be consumed. This paper discusses in this context some major quality characteristics of SOA systems and new measurement approaches, considering complex IT systems and digital ecosystems. This approach is based on growth models related to SOA's reference architecture as well as diversity and entropy. On a field test certain metrics are applied to a real large scale SOA system. As a result some major and important characteristics of the SOA system can be proven and weak structures in services orchestrations exposed.