[ZPK05b] Yi Zhang, David Parker and Marta Kwiatkowska. Grid-enabled Probabilistic Model Checking with PRISM. In Proc. 4th All Hands Meeting Workshop (AHM'05). September 2005. [ps.gz] [pdf] [bib]
Downloads:  ps.gz ps.gz (32 KB)  pdf pdf (58 KB)  bib bib
Abstract. In this paper, we present our work on extending the probabilistic model checking tool PRISM with Grid computing capabilities. PRISM is a tool for the verification and analysis of large, probabilistic models, a process which involves numerical solution techniques. Like most verification tools, PRISM can be very resource intensive and will benefit greatly from parallelisation efforts. Grid computing provides an invaluable means through which to invoke parallel numerical solution engines in PRISM, providing support for job distribution, data transfer and task monitoring on large-scale parallel and distributed computing environments. We describe the middleware we are developing to integrate this functionality into PRISM and illustrate its use on large model checking case study.