| M. Hennessy. Concurrent testing of processes. Acta Informatica, 32, 509--543, |
....verification techniques and tools such as model checkers. However, as is well known, the state machine approach is inherently inadequate for describing action refinement, the operation of refining atomic actions in a concurrent process, which suggests aspects of top down modular development [1, 2, 3, 10, 16, 20, 32, 39, 40, 41, 47] and changes of granularity [28, 31] This limitation is a direct result of the identification of concurrent actions with sequential, interleaved actions. For example, the state machine representation of the concurrent process a jj b, which can concurrently perform an a and b action, is ....
....our methods to handle these cases. Other Results, Open Problems, and Future Work There is not yet a consensus on what an action refinement operator should be. For example, our action refinement operator and that of [47] are tuned to a CSP style synchronizationwith restriction, while those of [3, 20] are tuned to a CCS style synchronization by hidingcomplementary actions. In this regard, an action refinement theory closely related to ours has been proposed by Hennessy [20] His theory incorporates an interesting, and in certain respects more powerful, action refinement operation, and he has ....
[Article contains additional citation context not shown here]
M. C. Hennessy. Concurrent testing of processes. In Proceedings of CONCUR '92, Volume 630 of Lecture Notes in Computer Science, pages 94--107, 1992.
....In these tools, the expected result of each test case usually is specified by a table. For parallel and distributed reactive systems this is not appropriate, since the correctness not only depends on the functional value, but also on the relative order of events. Recently, a number of researchers [12, 4, 13] have proposed testing based on formal specifications. In this approach, for each test case the result of the program run is checked whether it matches a formal description in some specification language. The result of a reactive program run often is modelled as a finite or infinite sequence of ....
M. Hennessy. Concurrent testing of processes. Acta Informatica, 32:509--543, 1995.
....some form of test. At least, we would need more than a sequential test. For a discussion of this problem for the case of pomset trace equivalence see [14] Concerning other related work, there are a number of approaches to define testing equivalences which are invariant under action refinement [21, 11, 12, 3]. However, all these approaches use the idea of split or ST equivalences: It is taken into account that actions have a duration, however they do not take causalities precisely into account (they will in general not respect pomset trace equivalence) Due to lack of space, all proofs have been ....
M. Hennessy. Concurrent testing of processes. In Cleaveland [5], pages 94--107.
....be noted that action a has not yet started its execution in state S 1 , that it is being executed in state S 2 , and finally that it has terminated its execution in state S 3 . The ST semantics, using the splitting technique introduced above, has been applied in the literature to process algebras [1, 9]; such an approach is however not completely satisfactory, as the splitting of the actions may significantly contribute to the classical state space explosion problem. In [18] an alternative definition of the ST bisimulation has been proposed for prime event structures with silent moves; the main ....
.... defined for Petri nets [17] and later translated to other models, like the prime event structures [16, 18] For prime event structures, the ST bisimulation has been proven to be the coarsest relation preserved under action refinement [18] and it has also been applied to concrete process algebra [1, 9]. The key point of the ST semantics consists in splitting each action into its start and its end, as illustrated below. Let a and a denote the start and the end of action a respectively. Previous behaviors P and Q may then be written as: P j a; a; stopjjjb; b; stop Q j a; a; b; b; stop[ b; b; a; ....
M. Hennessy, Concurrent Testing of Processes. In proc. of CONCUR'92, Volume 630 of LNCS, pages 94-107, Springer-Verlag, 1992.
....author of the present paper this fact has so far not been explicitly stated in the context of De Nicola Hennessy style testing. On the other hand it has been shown that the combination of action refinement and testing too requires a distinction like the above (see, for example, Vog89] JM92] [Hen92]) To show that the rendezvous can be understood in terms of action refinement seems to be an open problem. 5. The Pure Rendezvous Calculus as an Extension of CCS without s From a practical point of view one expects that a certain restricted rendezvous discipline has the same power as the ....
M. Hennessy. Concurrent Testing of Processes. In W.R. Cleaveland, editor, CONCUR '92, LNCS 630, pages 94--107. Springer, 1992.
....holds in our theory: SC(P 1 ) j SC(P 2 ) SC(P 1 jj P 2 ) In a CCS style setting, processes are generally already self synchronized on complementary actions, that is, they satisfy the identity SC(Q) Q. For example, the CCS like processes considered by Aceto Hennessy in [1] and Hennessy in [8] are all self synchronized in this way. For such processes, we then have Q 1 j Q 2 = SC(Q 1 jj Q 2 ) Similar identities hold for SCCS style [11] and Meijestyle [3] synchronized parallel composition. The intuitive simplicity of self synchronization is reflected in various algebraic identities ....
....operator and is the CCS choice operator. 1.2 Self Synchronization and Action Refinement We have found self synchronization particularly helpful in clarifying the relationship between contrasting notions of action refinement proposed in the literature. Aceto Hennessy [1] and Hennessy [8] have developed a CCS like process language with an action refinement operator based on syntactic substitution, in which communication can occur between refining processes. Hennessy [8] suggests that such an action refinement with communication may be more useful than the non communicating version ....
[Article contains additional citation context not shown here]
M. C. Hennessy. Concurrent testing of processes. In Proceedings of CONCUR '92, Volume 630 of Lecture Notes in Computer Science, pages 94-- 107, 1992.
....is obtained as the limit of split n failure semantics. AH c, Aceto a] define refine equivalence, also on a syntactical level, but on all of CCS, and prove that it is preserved under syntactic action refinement. This equivalence coincides (at least for free systems) with STb . Likewise, [Hennessy b] defines an ST testing equivalence on CCS and proves preservation under syntactic refinement. This equivalence coincides with STf . Finally, we would like to address the question whether history preserving bisimulation as defined here is the coarsest equivalence finer than pomset bisimulation ....
M. Hennessy (1995): Concurrent testing of processes. Acta Informatica 32(6), pp. 509--543.
....as well as some of the basic ideas of pomset runs and causal partial orders of Net theory. 1 Introduction The operation of refining atomic actions in a concurrent process suggests aspects of top down modular development and also requires use of some sort of true concurrency process model [1, 2, 3, 5, 8, 12, 13, 14, 17]. In a previous paper [10] we developed a semantics, Delta] MUST , for certain simple splitting and choice action refinements on a Petri Net model of processes. Our semantics generalizes and simplifies a similar semantics developed in a seminal paper by Vogler [17] These semantics ....
....a complete axiom system for equations between closed recursion free CSP CCS process terms corresponding to the (non divergent) isolated elements in our semantical spaces. There is not yet a consensus on what an action refinement operator should be. For example, the action refinement operator of [3, 8, 14, 17] contrasts with the one used in [1, 2] since the operators of [3, 8, 14, 17] distribute over CCS choice but not over a Hoare Hennessy external choice operator, H , while the operators of [1, 2] distribute over H but not over CCS choice. While our [ Delta] MUST intvl semantics is ....
[Article contains additional citation context not shown here]
M. C. Hennessy. Concurrent testing of processes. In Proceedings of 3 rd CONCUR, 1992. Appears in this volume.
....distinctions can be made at the level of syntax, without resorting to reasoning about event occurrences. 1 Introduction This paper is part of an effort by many authors to give a truly concurrent semantics to process algebra. Many authors have previously contributed in this area, see e.g. [22, 19, 9, 20, 4, 11, 15, 1, 17, 2], and the methods have included translation into Petri nets, deriving an event structure, and enhancing the labels in labelled transition systems, often leading to descriptions of lowlevel character. We propose starting at the high level of abstraction usually associated with process algebras, and ....
M. Hennessy. Concurrent testing of processes. Technical Report 11/91, University of Sussex, 1991.
....distinctions can be made at the level of syntax, without resorting to reasoning about event occurrences. 1 Introduction This paper is part of an effort by many authors to give a truly concurrent semantics to process algebra. Many authors have previously contributed in this area, see e.g. [20, 17, 8, 18, 4, 9, 13, 1, 15, 2], and the methods have included translation into Petri nets, deriving an event structure, and enhancing the labels in labelled transition systems, often leading to descriptions of lowlevel character. We propose starting at the high level of abstraction usually associated with process algebras, and ....
M. Hennessy. Concurrent testing of processes. Technical Report 11/91, University of Sussex, 1991.
No context found.
M. Hennessy. Concurrent testing of processes. Acta Informatica, 32, 509--543,
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