Improving Ariadne's Bundle by Following Multiple Threads in Abstraction Refinement

We propose an abstraction refinement method for invariant checking,which is based on the simultaneous analysis of all abstractcounter examples of shortest length in the current abstraction. Thealgorithm is focused on an improved Ariadne's Bundle of SORs(Synchronous Onion Rings) of the abstract...

Full description

Saved in:
Bibliographic Details
Published in:International Conference on Computer Aided Design: Proceedings of the 2003 IEEE/ACM international conference on Computer-aided design; 09-13 Nov. 2003 p. 408
Main Authors: Wang, Chao, Li, Bing, Jin, HoonSang, Hachtel, Gary D., Somenzi, Fabio
Format: Conference Proceeding
Language:English
Published: Washington, DC, USA IEEE Computer Society 09.11.2003
Series:ACM Conferences
Subjects:
Theory of computation > Logic > Programming logic
Theory of computation > Semantics and reasoning > Program reasoning
Theory of computation > Semantics and reasoning > Program reasoning > Invariants
Theory of computation > Semantics and reasoning > Program reasoning > Program verification
ISBN:9781581137620, 1581137621
ISSN:1092-3152
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose an abstraction refinement method for invariant checking,which is based on the simultaneous analysis of all abstractcounter examples of shortest length in the current abstraction. Thealgorithm is focused on an improved Ariadne's Bundle of SORs(Synchronous Onion Rings) of the abstract model; the transitionsthrough these SORs contain all shortest ACEs (Abstract CounterExamples) and no other ACEs. The SORs are exploited in twodistinct ways to provide global guidance to the abstraction refinementprocess: (1) Refinement variable selection is based on theentirety of transitions connecting the SORs, and (2) a SAT-basedconcretization test is formulated to test all ACEs in the SORs atonce. We call this test multi-thread concretization. The scalabilityof our refinement algorithm is ensured in the sense that all theanalysis and computation required in our refinement algorithm areconducted on the abstract model.The abstraction efficiency of a given abstraction refinement algorithmmeasures how much of the concrete model is required tomake the decision. We include experimental comparisons of ournew method with recently published techniques. The resultsshow that our scalable method, based on global guidance from theentire bundle of shortest ACEs, outperforms these other methods interms of both run time and abstraction efficiency.
Bibliography:SourceType-Conference Papers & Proceedings-1
ObjectType-Conference Paper-1
content type line 25
ISBN:9781581137620
1581137621
ISSN:1092-3152
DOI:10.5555/996070.1009923