D. A. Peled. Ten years of partial order reduction. In Computer-Aided Verification (CAV), volume 1427 of Lecture Notes in Computer Science (LNCS), pages 17--28. Springer-Verlag, 1998.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....importent pour l evaluation de la formule. Cet ensemble dit observable est not e AP obs . Par cons equent on peut associer un sous ensemble de variables atomiques AP obs , un sous ensemble de transitions dites observables TObs . 9 D efinition 14. Transition Observable par rapport a AP obs [Pel98] Gamma Gamma g t2T ; i une SKE homog ene et AP obs AP . Une transition t 2 T est dite observable par rapport a AP obs ssi 9s 1 ; s 2 2 S : s 1 Gamma Gamma s 2 et obs (s 1 ) 6= obs (s 2 ) o u obs : S 7 AP est d efinie par obs (s) s) AP obs L ensemble des transitions ....

....de pas couvrant introduits dans [VAM] pour pr eserver des propri et es sp ecifiques exprim ees en logique temporelle lin eaire LTL GammaX . Le travail en cours consiste a comparer le b en efice de la r eduction obtenue, avec ceux obtenus par d autres m ethodes telles que les ensembles amples [Pel98]. Pour cela, nous allons impl ementer ces deux m ethodes dans l environnement TINA (TImed Net Analyser http: www.laas.fr tina ) afin de pouvoir les exp erimenter. 17 ....

D. Peled. Ten years of partial order reduction. In Proceedings of CAV. Springer Verlag, LNCS 1427, 1998.

....of many verification methods for concurrent systems. The state space is then analyzed by available efficient and automatic verification techniques, such as bisimulation and model checking. The combinatorial explosion is the main limitation of these approaches. The partial order techniques (see [GW93,Pel98] for a survey) are the framework of the approach developed in this paper. Their basic principle is to consider a single specific path among all the sequences which possess the same Mazurkiewicz trace [Maz86] In the case of persistent sets [WG93] only a subset of enabled transitions is examined, ....

....preservation of other classes of properties. For example the ignoring problem is an advantage for deadlock detection, but must be avoid for more general safety properties[God96] Prospective work concerns the preservation of specific properties (safety and liveness) including a study of ample set [Pel98] and a comparison of trace automata obtained by persistent set and CSG. ....

Doron Peled. Ten years of partial order reduction. In Proceedings of CAV'98, pages 17--28. Springer Verlag, LNCS 1427, 1998.

....therefore seems to circumvent the state space explosion problem. At least, our experience has been that MFGs describing a system s behavior are often smaller than the full state graph for the system. Furthermore, MFGs are often smaller than reduced state graphs computed using partial order methods [Pel98]. This is illustrated in Section 3.3. One could regard our framework as having a powerful built in partial order method, because it does not consider interleavings of messages sent (or received) by a component on di#erent channels. Such inter channel orderings cannot be represented directly in ....

Doron Peled. Ten years of partial order reduction. In Alan J. Hu and Moshe Y. Vardi, editors, Proc. 10th Int'l. Conference on Computer-Aided Verification (CAV), volume 1427 of Lecture Notes in Computer Science, pages 17--28. Springer-Verlag, 1998.

....these structures for modeling concurrent systems. The exciting point on Mazurkiewicz traces is that only a single representative for each equivalence class might be employed for verifying a desired property. This is the insight underlying many of the partial order based verification methods (e.g. Pel98, Val91] As may be guessed, the importance of these methods lies in the fact that the computational resources required for the verification task can often be dramatically reduced. In general, two approaches for the verification of systems can be distinguished: model checking and theorem proving ....

....It turns out that these procedures are as e#cient as for LTL (for words) viz they are exponential in the length of the formula and linear in the size of the system and are essentially optimal. The model checking procedure employs an optimization which is similar to partial order reduction [Pel98] and SCCS [Mil83] but is adapted to the special nature of our underlying systems. Outline of this thesis In the next chapter, we describe a typical scenario of a distributed system. Here we will spot the characteristics of such a system. This will guide us to define the formal notion of ....

[Article contains additional citation context not shown here]

D. Peled. Ten years of partial order reduction. In Proceedings of 10th International Conference on Computer-Aided Verification (CAV'98), volume 1427 of Lecture Notes in Computer Science, pages 17--28, Vancouver, BC, Canada, 1998. Springer.

....our compiler generator to support the distributed functional programming language Erlang. 4 Truth: The Parallel Version Despite the improvements of model checking techniques during the last years, the so called state space explosion still limits its application. While partial order reduction [14] or symbolic model checking [10] reduce the state space by orders of magnitude, typical veri cation tasks still last days on a single workstation or are even (practically) undecidable due to memory restrictions. On the other hand, cheap yet powerful parallel computers can be constructed by ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Veri cation, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....of the resulting state space, known as the state explosion problem. State space explosion occurs due to non determinism in the model introduced by data or concurrency. Different approaches have been proposed to tackle this problem. One of the most successful techniques is partial order reduction [19]. This method explores a reduced state space by exploiting the independence of concurrently executed events. Partial order reduction is particularly efficientinasynchronous systems, where manyinterleavings of events are equivalent with respect to a given property specification. Considering only ....

D. Peled. Ten years of partial order reduction. In Computer Aided Verification, number 1427 in Lecture Notes in Computer Science, pages 17--28. Springer, 1998.

....method is the size of the resulting state space, known as the state explosion problem. It occurs due to non determinism in the model introduced by data or concurrency. Different approaches have been proposed to tackle this problem. One of the most successful techniques is partial order reduction [22]. This method explores a reduced state space by exploiting the independence of concurrently executed events. Partial order reduction is particularly efficient in asynchronous systems, where many interleavings of concurrent events are equivalent with respect to a given property specification. ....

D. A. Peled. Ten years of partial order reduction. In Computer Aided Verification, number 1427 in Lecture Notes in Computer Science, pages 17--28. Springer, 1998.

....one of these n sequences of n transitions is generated from G to G 0 . Earlier partial order reduction methods were developed for verifying deadlock freedom and other safety properties, while recent partial order reduction methods provide solutions to the model checking problem [3] 4] and [11] . Simultaneous reachability analysis (SRA) is a recently proposed approach to alleviating the state space explosion problem. SRA differs from partial order reduction in that the former allows a global transition in a reachability graph to contain a set of independent local transitions. Consider ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Verification, number 1427 in LNCS, pages 17--28, Vancouver, BC, Canada, 1998. Springer.

....the resulting state space, known as the state explosion problem. State space explosion occurs due to non determinism in the model introduced by data or concurrency. 1 Different approaches have been proposed to tackle this problem. One of the most successful techniques is partial order reduction [19]. This method explores a reduced state space by exploiting the independence of concurrently executed events. Partial order reduction is particularly efficient in asynchronous systems, where many interleavings of events are equivalent with respect to a given property specification. Considering only ....

D. Peled. Ten years of partial order reduction. In Computer Aided Verification, number 1427 in Lecture Notes in Computer Science, pages 17--28. Springer, 1998.

....property. Numerous case studies have shown that this approach improves the early detection of errors during the design process (see [8] for an overview) Despite the developments in the last years, the so called state space explosion still limits its application. While partial order reduction [32] or symbolic model checking [28] reduce the state space by orders of magnitude, typical veri cation tasks still last days on a single workstation or are even (practically) undecidable due to memory restrictions (see for example [15] On the other hand, cheap yet powerful parallel computers can ....

D. Peled. Ten years of partial order reduction. In CAV, Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....not to distinguish models from one and the same equivalence class (i.e. which are linearisations of one and the same partial order) such that one has to check a speci ed property for just one member of each equivalence class. Thus, it is possible to employ so called partial order reduction methods [19, 24]. In Section 2, we give a formal de nition of our MSCs, channels and message passing automata. Section 3 presents our linear time temporal logic which is useful for specifying sets of MSCs. Recalling the notion of alternating B uchi automata in Section 4, we are ready to present our main procedure ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....whether the (transition) system satis es this property. Numerous case studies have shown that this approach improves the early detection of errors [3] Despite the developments in the last years, the so called state space explosion still limits its application. While partial order reduction [20] or symbolic model checking [17] reduce the state space by orders of magnitude, typical veri cation tasks still last days on a single workstation or are even (practically) undecidable due to memory restrictions (see for example [7] On the other hand, cheap yet powerful parallel computers can be ....

....is already of practical interest since it allows to check larger systems, it should also be considered as a further attempt to develop parallel model checking algorithms. More research should be carried out in this direction. Especially, on the y model checking and partial order reduction [20] should be analysed with respect to parallelisation. Furthermore, di erent (especially non P complete) speci cation logics might provide better parallel model checking algorithms. ....

D. Peled. Ten years of partial order reduction. In CAV, Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....of computations will have the desired property or none will do ( leads to deadlock is one such property) For verifying such properties one has to check them for just one member of each equivalence class. This is the insight underlying many of the partial order based veri cation methods (e.g. [Pel98,Val91]) As may be guessed, the importance of these methods lies in the fact that via these methods the computational resources required for the veri cation task can often be dramatically reduced. It is often the case that the equivalence classes of computations generated by a distributed system ....

....LTL over Mazurkiewicz traces. Given an LTL formula over Mazurkiewicz traces, consider its automaton A . It is a linear trace consistent automaton over words. For this kind automata, several powerful partial order reduction techniques have been developed, which will have the same success here [Pel98]. Hence, specifying with LTL over Mazurkiewicz traces promises despite the bad worst case runtime of its decision procedure ecient veri cation tasks in practice. Note that the rst item even implies that the languages de nable by LTLformulas over Mazurkiewicz traces are FO de nable over ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Veri- cation, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....out that these procedures meet the known complexity bounds for LTL viz exponential in the length of the formula and linear in the size of the system and are essentially optimal. The model checking procedure employs an optimisation which is similar to a technique known as partial order reduction [Pel98]. However, instead of de ning the interleaving product of the sequential processes and then trying to omit states not in uencing the result of the model checking procedure, we are able to de ne smaller systems directly because of the underlying model. Synchronous systems were studied by several ....

....one linearisation. For example, BA for A as in Figure 4 is shown in Figure 6 Complexity Model Checking is exponential in the size of the formula and linear in the size of BA . The size of BA is exponential in the size number of components of A. The experiences gained by partial order reduction [Pel98] allow the conclusion that the number of reachable states is in the average case much smaller. 9 Conclusion In this paper we presented a framework for verifying synchronous hardware systems. We introduced a suitable semantic model, Distributed Synchronous Transition Systems (DSTS) which are ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

....represent nite state reactive systems. The use of bdds allows for the veri cation of systems with more than 10 20 states. However, the state explosion problem (due to the complexity of the design) remains the major bottleneck of this approach. Many reduction techniques have been developed [11, 27, 14, 22] to attack the state explosion problem. The most widely used and promising technique is Abstraction [11] The intuitive idea behind Abstraction is to abstract a model to a possibly small model, such that if a speci c property holds for the abstract model, the property holds for the original model ....

Doron Peled. Ten years of partial order reduction. In Proc. Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer-Verlag.

....notion of independence working together concurrently. Traditionally, these are transformed into a single system by interleaving. However, this causes an exponential blow up in the number of components, known as state space explosion. Besides symbolic model checking [McM93] partial order reduction [Pel98] is a powerful approach to avoid this explosion. Its idea is to simplify the resulting system by employing its inherent independence. Of course, partial order reduction is only applicable for speci cations that respect this independence. However, to decide whether a formula is so called trace ....

....this explosion. Its idea is to simplify the resulting system by employing its inherent independence. Of course, partial order reduction is only applicable for speci cations that respect this independence. However, to decide whether a formula is so called trace consistent is complete in PSPACE [Pel98]. An approach on the same line is to describe the runs of a system by partial orders which are Mazurkiewicz traces [Maz77, DR95] and to formulate speci cations directly for traces. This guarantees that every formula of a corresponding logic is automatically trace consistent. Furthermore, given a ....

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Verication, number 1427 in LNCS, pages 17-28, Vancouver, BC, Canada, 1998. Springer.

No context found.

D. A. Peled. Ten years of partial order reduction. In Computer-Aided Verification (CAV), volume 1427 of Lecture Notes in Computer Science (LNCS), pages 17--28. Springer-Verlag, 1998.

No context found.

D. A. Peled. Ten years of partial order reduction. In Computer-Aided Verification (CAV), volume 1427, pages 17--28, 1998.

No context found.

Peled DA (1998) Ten years of partial order reduction. In: 10th conference on computer-aided verification (CAV). Lecture notes in computer science, vol 1427. Springer, Berlin Heidelberg New York, pp 17--28

No context found.

Peled, D.: Ten years of partial order reduction. In: Proceedings of the 10th International Conference on Computer Aided Verification, Springer-Verlag (1998) 17--28

No context found.

D. Peled. Ten years of partial order reduction. In Computer Aided Verification, number 1427 in Lecture Notes in Computer Science, pages 17--28. Springer, 1998.

No context found.

D. A. Peled. Ten years of partial order reduction. In 10th International Conference on Computer-Aided Verification (CAV), number 1427 in Lecture Notes in Computer Science 1427, pages 17--28. Springer, 1998.

No context found.

D.A. Peled. Ten years of partial order reduction. In CAV, Computer Aided Verification, number 1427 in LNCS, pp. 17--28, Vancouver, BC, Canada, 1998. Springer.

No context found.

D.A. Peled. Ten years of partial order reduction. In CAV, Computer Aided Verification, number 1427 in LNCS, pp. 17--28, Vancouver, BC, Canada, 1998. Springer.

No context found.

Doron Peled. Ten years of partial order reduction. In CAV, Computer Aided Veri#cation,number 1427 in LNCS, pages 17#28, Vancouver, BC, Canada, 1998. Springer.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC