A New Way of Automating Statistical Testing Methods
Proceedings of the 16th IEEE international conference on Automated software engineering
A Generic Method for Statistical Testing
ISSRE '04 Proceedings of the 15th International Symposium on Software Reliability Engineering
Uniform random sampling of traces in very large models
Proceedings of the 1st international workshop on Random testing
Coverage-biased Random Exploration of Models
Electronic Notes in Theoretical Computer Science (ENTCS)
Uniform Monte-Carlo model checking
FASE'11/ETAPS'11 Proceedings of the 14th international conference on Fundamental approaches to software engineering: part of the joint European conferences on theory and practice of software
ICTSS'11 Proceedings of the 23rd IFIP WG 6.1 international conference on Testing software and systems
| Hi-index | 0.00 |
This paper presents first experimental results on uniform random generation of paths in very large graphs that model concurrent systems, as it was presented in [3]. The approach is based on techniques and tools for counting and drawing uniformly at random in combinatorial structures. It exploits the fact that in a system made of several concurrent components, local uniform drawings of component paths can be combined into a very good approximation of uniform drawing of paths in the global system, without constructing the global model. The paper describes some implementation of the methods presented in [3], reports results on a first suite of benchmarks, exploring the limits and the possibility of this new approach to uniform random walks.