Skip to Main content Skip to Navigation

Combining Parallel Emptiness Checks with Partial Order Reductions

Abstract : In explicit state model checking of concurrent systems, multi-core emptiness checks and partial order reductions (POR) are two major techniques to handle large state spaces. The first one tries to take advantage of multi-core architectures while the second one may decrease exponentially the size of the state space to explore. For checking LTL properties, Bloemen and van de Pol [2] shown that the best performance is currently obtained using their multi-core SCC-based emptiness check. However, combining the latest SCC-based algorithm with POR is not trivial since a condition on cycles, the proviso, must be enforced on an algorithm which processes collaboratively cycles. In this paper, we suggest a pessimistic approach to tackle this problem for liveness properties. For safety ones, we propose an algorithm which takes benefit from the information computed by the SCC-based algorithm. We also present new parallel provisos for both safety and liveness properties that relies on other multi-core emptiness checks. We observe that all presented algorithms maintain good reductions and scalability.
Document type :
Conference papers
Complete list of metadatas

Cited literature [28 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-02408440
Contributor : Denis Poitrenaud <>
Submitted on : Friday, December 13, 2019 - 8:49:34 AM
Last modification on : Friday, March 27, 2020 - 2:27:59 AM
Document(s) archivé(s) le : Saturday, March 14, 2020 - 1:28:50 PM

File

 Restricted access
To satisfy the distribution rights of the publisher, the document is embargoed until : 2020-06-13

Please log in to resquest access to the document

Identifiers

Citation

Denis Poitrenaud, Etienne Renault. Combining Parallel Emptiness Checks with Partial Order Reductions. ICFEM 2019 - 21st International Conference on Formal Engineering Methods, Nov 2019, Shenzhen, China. pp.370-386, ⟨10.1007/978-3-030-32409-4_23⟩. ⟨hal-02408440⟩

Share

Metrics

Record views

88