Checking Deadlock-Freedom of Parametric Component-Based Systems

Bozga, Marius; Iosif, Radu; Sifakis, Joseph

doi:10.1007/978-3-030-17465-1_1

Cited by 11 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Usually, computing invariants requires building a sequence of underapproximants whose limit is the least fixed point of an abstraction of the transition relation of the system [20]. This is not the case of the trap invariant, that can be directly computed from the trap constraint and the initial marking [11,15].…”

Section: Trap Invariantsmentioning

confidence: 99%

“…Grenoble Alpes 3 We use the term architecture for the shape of the graph along which the interactions take place. rameterized systems in [15], but the work there is only applicable to clique architectures, in which processes are indistinguishable, and the system can be described by one single Petri Net with an infinite family of initial markings. Here, for the first time, we show that the trap technique can be extended to pipelines, token rings and trees, where the system is defined by an infinite family of Petri Nets, each with a different structure.…”

Section: Introductionmentioning

confidence: 99%

“…Such invariants depend only on the structure (and not on the operational semantics) of an infinite family of Petri Nets, one for each instance of the system, being thus structural invariants. Essentially, the invariants we infer use the traps 4 of the system, which are sets W of local states with the property that, if a process is initially in a state from W, then always some process will be in a state from W. Following [11,15], we call them (parameterized) trap invariants. Computing trap invariants only requires a simple syntactic transformation of the interaction formula and the result is expressed using WSκS, the weak monadic second order logic of κ ≥ 1 successor functions.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Structural Invariants for the Verification of Systems with Parameterized Architectures

Bozga

Esparza

Iosif

et al. 2020

Tools and Algorithms for the Construction and Analysis of Systems

Self Cite

View full text Add to dashboard Cite

We consider parameterized concurrent systems consisting of a finite but unknown number of components, obtained by replicating a given set of finite state automata. Components communicate by executing atomic interactions whose participants update their states simultaneously. We introduce an interaction logic to specify both the type of interactions (e.g. rendez-vous, broadcast) and the topology of the system (e.g. pipeline, ring). The logic can be easily embedded in monadic second order logic of κ ≥ 1 successors (WSκS), and is therefore decidable.Proving safety properties of such a parameterized system, like deadlock freedom or mutual exclusion, requires to infer an inductive invariant that contains all reachable states of all system instances, and no unsafe state. We present a method to automatically synthesize inductive invariants directly from the formula describing the interactions, without costly fixed point iterations. We experimentally prove that this invariant is strong enough to verify safety properties of a large number of systems including textbook examples (dining philosophers, synchronization schemes), classical mutual exclusion algorithms, cache-coherence protocols and self-stabilization algorithms, for an arbitrary number of components. arXiv:2002.07672v1 [cs.DC] 18 Feb 2020correctness for at most c processes implies correctness for any number of processes [16,25,24,7,31]. Other methods identify systems with well-structured transition relations [29,1,27]. An exhaustive chart of decidability results for verification of parameterized systems is drawn in [12]. When decidability is not of concern, over-approximating and semi-algorithmic techniques such as regular model checking [33,2], SMT-based bounded model checking [4,19], abstraction [10,14] and automata learning [17] can be used to deal with more general classes of systems.The efficiency of a verification method crucially relies on its ability to synthesize an inductive safety invariant, i.e., an infinite set of configurations that contains the initial configurations, is closed under the transition relation, and excludes the error configurations. In general, automatically synthesizing invariants requires computationally expensive fixpoint iterations [20]. In the particular case of parameterized systems, invariants can be either global, relating the local states of all processes [21], or modular, relating the local states of a few processes whose identity is irrelevant [35,18]. Our Contributions. The novelty of the approach described in this paper is three-fold:1. The architecture of the system is not fixed a priori, but given as a parameter of the verification problem. In fact, we describe parameterized systems using the Behavior-Interaction-Priorities (BIP) framework [9], in which processes are instances of finite-state component types, whose interfaces are sets of ports, labeling transitions between local states, and interactions are sets of strongly synchronizing ports, described by formulae of an interaction logic. An interaction formula captur...

show abstract

Section: Trap Invariantsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Structural Invariants for the Verification of Systems with Parameterized Architectures

Bozga

Esparza

Iosif

et al. 2020

Tools and Algorithms for the Construction and Analysis of Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…To verify correctness of the parametric interacting system of a motif we extend the approach adopted for static BIP: assuming that dynamic connectors correctly enforce the sought coordination, it remains to show that restricting the behavior of deadlock-free components does not introduce deadlocks. We have recently shown this approach can be extended for parametric systems [5].…”

Section: Discussionmentioning

confidence: 99%

Programming Dynamic Reconfigurable Systems

Ballouli

Bensalem

Bozga

et al. 2018

Formal Aspects of Component Software

Self Cite

View full text Add to dashboard Cite

DR-BIP is an extension of the BIP component framework intended for programming reconfigurable systems encompassing various aspects of dynamism. It relies on architectural motifs to structure the architecture of a system and to coordinate its reconfiguration at runtime. An architectural motif defines a set of interacting components that evolve according to reconfiguration rules. With DR-BIP, the dynamism can be captured as the interplay of dynamic changes in three independent directions 1) the organization of interactions between instances of components in a given configuration; 2) the reconfiguration mechanisms allowing creation/deletion of components and management of their interaction according to a given architectural motif; 3) the migration of components between predefined architectural motifs which characterizes dynamic execution environments. The paper lays down the formal foundation of DR-BIP, illustrates its expressiveness on few examples and discusses avenues for dynamic reconfigurable system design.

show abstract

“…In order to guarantee the deadlock freedom, two methods are provided: Invariant preservation of one additional invariant, and model checking through animation (Bozga et al 2019). The first method is based on adding an invariant that include the disjunction of all the model events guards.…”

Section: Verification Using Proof Obligationsmentioning

confidence: 99%

Formal approach to model complex adaptive computing systems

Jarrar¹,

Wakrime

Balouki³

2020

Complex Adapt Syst Model

View full text Add to dashboard Cite

Complex adaptive systems provide a significant number of concepts such as reaction, interaction, adaptation, and evolution. In general, these concepts are modelled employing different techniques which give an inexplicit vision on the system. Therefore, all concepts must be carefully modelled using the same approach to avoid contradiction and guarantee system homogeneity and correctness. However, developing a computing system that includes all these concepts using the same approach is not an easy task and requires a perfect understanding of the system’s behaviour. In this paper, we contribute as stepwise towards proposing an approach to model the most important concepts of complex adaptive systems while ensuring homogeneity and the correctness of models. For this aim, we present five standard agent-based models formalizing agent properties, reaction, interaction, adaptation, and evolution. These models are adapted to all cases of complex adaptive systems since they include an abstract description of these concepts. To implement our approach formally, we choose the Event-B method due to the strong assurance of bugs’ absence that it guarantees. Besides, it supports horizontal and vertical refinement which facilitates the specification process. Furthermore, the approach of this paper addresses the very abstract level of modelling which expand the use of this approach to other formal methods and tools.

show abstract

Checking Deadlock-Freedom of Parametric Component-Based Systems

Cited by 11 publications

References 22 publications

Structural Invariants for the Verification of Systems with Parameterized Architectures

Structural Invariants for the Verification of Systems with Parameterized Architectures

Programming Dynamic Reconfigurable Systems

Formal approach to model complex adaptive computing systems

Contact Info

Product

Resources

About