Evaluation and extracting factual software architecture of distributed system by process mining techniques

Abstract

The factual software architectures that are actually implemented of distributed systems do not conform the planned software architectures (Beck 2010). It happens due to the complexity of distributed systems. This problem begets two main challenges; First, how to extract the factual software architectures with the proper techniques and second, how to compare the planned software architecture with the extracted factual architecture. This study aims to use process mining to discover factual software architecture from codes and represents software architecture model in Petri Net to evaluate model by the linear temporal logic and process mining. In this paper, the applicability of process mining techniques, implemented in the ProM6.7 framework is shown to extract and evaluate factual software architectures. Furthermore, capabilities of Hierarchical Colored Petri Net implemented in CPN4.0 are exploited to model and simulate software architectures. The proposed approach has been conducted on a case study to indicate applicability of the approach in the distributed data base system. The final result of the case study indicates process mining is able to extract factual software architectures and also to check its conformance.