In this paper we will describe the SoftArc approach. With the SoftArc approach it is possible to model and analyse safety-critical embedded and distributed systems that consist of both hard- and software. We are going to present the SoftArc modelling language, its syntax and semantics. The semantics of the SoftArc modelling language is defined in terms of stochastic reactive modules. We will show how important measures of interest for probabilistic dependability analysis like availability, unavailability, and survivability, can be analysed. We will demonstrate the feasibility of our approach by means of two case studies, that involve hard- and software elements. First, we are presenting two industrial case studies from the automotive industry. We will analyse the non volatile random access manager (NVRAM) from the AUTOSAR open system architecture, Second, we are going to present the survivability analysis of a simplified version of the Google replicated file system.
|Title of host publication||Proceedings of the ESREL 2010 Annual Conference|
|Editors||Kent Bladh, Jan-Erik Holmberg, Johanna Oxstrand, Pekka Pyy|
|Place of Publication||London|
|Publisher||Taylor & Francis|
|Number of pages||9|
|Publication status||Published - 5 Sep 2010|
|Event||European Safety and Reliability Conference 2010 : Reliability, Risk and Safety: Back to the Future - Rhodes, Greece|
Duration: 5 Sep 2010 → 9 Sep 2010
|Conference||European Safety and Reliability Conference 2010|
|Abbreviated title||ESREL 2010|
|Period||5/09/10 → 9/09/10|
Haverkort, B. R. H. M., Kuntz, G. W. M., Leitner-Fischer, F., Remke, A. K. I., & Roolvink, S. (2010). Probabilistic verification of Architectural software models using SoftArc and Prism. In K. Bladh, J-E. Holmberg, J. Oxstrand, & P. Pyy (Eds.), Proceedings of the ESREL 2010 Annual Conference (pp. 852-860). Taylor & Francis.