W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. In 3rd IEEE Int. Comp. Perf. and Dependability Symp. (IPDS), pp. 228-237, 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....and dependability questions, there is an interest in the occurrence probability of certain state sequences. Stated di#erently, we would also like to be able to express measures that address the probability on particular paths through the CTMC. Except for the recent work by Obal and Sanders [54], we are not aware of suitable mechanisms to express such measures. In the sequel, we will specifically address this issue. 3 Formal verification with model checking Whereas performance and dependability evaluation focusses on answering questions concerning quantitative system issues, ....

....or require a manual tailoring of the model so as to address the right subsets of states. Below we will address a rigorous but flexible way of expressing performability measures. Finally, note that Obal and Sanders recently proposed a technique to specify so called path based reward variables [54] by which the specification of measures over state sequences becomes more convenient, because it avoids the manual tailoring of the model. In the context of the stochastic process algebra PEPA, Clark et al. recently proposed the use of a probabilistic modal logic to ease the specification of ....

W.D. Obal II and W.H. Sanders. State-space support for path-based reward variables. Performance Evaluation, 35: 233--251, 1999.

....reward) below a certain threshold. Such a measure has not been considered in the literature so far; 4] did not address computational procedures for full CSRL. The only other work we are aware of that allows for the specification of path based measures, has been reported by Obal and Sanders [21, 22]. Roughly speaking, they allow one to analyse more detailed path based behaviour than we do, because they employ automata of quite a general shape to collect steady state rewards. On the other hand, their approach neither supports nesting of path and state properties, nor time or reward ....

W.D. Obal, W.H. Sanders. State-space support for pathbased reward variables. Performance Evaluation, 35: 233-- 251, 1999.

....Most of these techniques assume a continuous time Markov chain (CTMC) as underlying stochastic process. While the analysis of CTMCs focuses mostly on transient state and steady state (i.e. long run) characteristics, the speci cation and analysis of path measures is a subject of growing interest [25]. The temporal logic CSL (Continuous Stochastic Logic) developed originally by Aziz et al. 2, 3] and extended by Baier et al. 7] provides a powerful means to specify path based as well as traditional state based measures on CTMCs in a concise, exible and unambiguous way. CSL is based on the ....

W.D. Obal II and W.H. Sanders. State-space support for path-based reward variables. Perf. Ev., 35: 233-251, 1999.

....that is very di#cult to specify directly is the expected amount of gain obtained from the system until a particular state is reached, provided that all paths to that state obey certain properties. Recently, Obal and Sanders have proposed a technique to specify so called path based reward variables [14] by which the specification of measures over state sequences becomes more convenient, because it avoids the manual tailoring of the model. In the context of the stochastic process algebra PEPA, Clark et al. recently proposed the use of a probabilistic modal logic to ease the specification of ....

W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. In 3rd IEEE Int. Comp. Perf. and Dependability Symp. (IPDS), pp. 228--237, 1998.

....that is very dicult to specify directly is the expected amount of gain obtained from the system until a particular state is reached, provided that all paths to that state obey certain properties. Recently, Obal and Sanders have proposed a technique to specify so called path based reward variables [14] by which the speci cation of measures over state sequences becomes more convenient, because it avoids the manual tailoring of the model. In the context of the stochastic process algebra PEPA, Clark et al. recently proposed the use of a probabilistic modal logic to ease the speci cation of reward ....

W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. In 3rd IEEE Int. Comp. Perf. and Dependability Symp. (IPDS), pp. 228-237, 1998.

No context found.

W. D. Obal II and W. H. Sanders, "State-space support for pathbased reward variables," in Proc. 3rd Ann. IEEE International Computer Performance and Dependability Symposium (IPDS '98), Durham, North Carolina, Sept. 1998, pp. 228--237.

....generalities. Groups Groups are a set of actions may coordinate in some ways. Preselection groups, 5] both variable and persistent, are supported. In addition, a new type of group called postselection is supported to mimic action case behavior in SANs. Reward Variables Work by Obal and Sanders [2] has defined a type of reward variable that embeds a state machine into the reward structure. Mobius uses this concept in defining a more general, state machine based reward variable concept. This way of specifying reward variables is very general and powerful, and encompasses previously known ....

W. D. Obal II and W. H. Sanders. State-space support for path-based reward variables. In Proceedings of the

....models yield a model composition graph, which is analyzed to find its automorphism group. This group is used to partition the state space into equivalence classes and directly generate a smaller state space. In subsequent work, we developed a new approach to reward variable specification [26] that allows one to define reward measures on sequences of events in the model, as captured in a path automaton specification. A path based reward variable can have different reward structures for each state of the path automaton. Another result of this work was a state space construction ....

....the specification of the system model and the path based reward variable. This new approach frees the modeler from the need to add additional complexity to a system model in order to support path based performance measures. The goal of this paper is to build upon our previous work, presented in [26] and [25] to develop specification and model construction techniques for measure adaptive state space construction. In this work, we combine our methods for path based reward variables with our work on detecting and exploiting model symmetry. This is the final step needed to separate the modeling ....

[Article contains additional citation context not shown here]

W. D. Obal II, W. H. Sanders, State-space support for path-based reward variables, Performance Evaluation 35 (1) (1999) 233--251.

....experience. To the extent possible, we have incorporated new research results into our successful modeling tool, UltraSAN. For a number of practical reasons, many of our recent research results have only been developed to a prototype, and are not generally available to other users (for example, [15, 16, 17, 27, 28]) In particular, we would have liked to develop a new graph composition formalism of [28] but found it difficult to include in UltraSAN because of the inherently closed nature of the software design. Mobius has been designed to include these modeling techniques, as well as to be extensible to ....

W. D. Obal II and W. H. Sanders. State-space support for path-based reward variables. In Proceedings of the 3rd Annual IEEE International Computer Performance and Dependability Symposium (IPDS '98), pages 228--237, Durham, North Carolina, USA, Sept. 1998.

No context found.

W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. In 3rd IEEE Int. Comp. Perf. and Dependability Symp. (IPDS), pp. 228-237, 1998.

No context found.

W.D. Obal II and W.H. Sanders. State-space support for path-based reward variables. Performance Evaluation, 35: 233--251, 1999.

No context found.

W.D. Obal, W.H. Sanders. State-space support for pathbased reward variables. Performance Evaluation, 35: 233-- 251, 1999.

No context found.

W. Obal and W. Sanders, "State-space support for pathbased reward variables," Perform. Eval., vol. 35, no. 3-4, pp. 233--251, 1999.

No context found.

W.D. Obal II and W.H. Sanders. State-space support for path-based reward variables. Perf. Ev., 35: 233-251, 1999.

No context found.

W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. In Proceedings of the Computer Performance and Dependability Symposium, 228--237, 1998.

No context found.

W.D. Obal, W.H. Sanders, State-space support for path-based reward variables, in: Proceedings of the Computer Performance and Dependability Symposium, 1998, pp. 228--237.

No context found.

W.D. Obal and W.H. Sanders. State-space support for path-based reward variables. Performance Evaluation, 35:233--251, 1999.

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