Duke University, Durham Petri-net based models have been extensively used for performance & performability modeling to analyze computer & communication systems [ 1 71. However, in the dependability modeling co-unity, Petri-net based models have received co~i&rably less attention [8 101. This paper describes a methodology to con

The notion of system “resilience” is receiving increased attention in domains ranging from safety-critical applications to ubiquitous computing. After reviewing how resilience has been defined in these contexts, we discuss roles that performability can play in both its definition and evaluation.

We propose, ifl this paper, a new algorithm to compute the performability distribution. Its computational complexity is polynomial and it deals only with nonnegative numbers bounded by one. This important property allows us to determine truncation steps and so to improve the execution time of the algorithm.

This extended abstract investigates the performability features of a car safety controller operating in a road tunnel. The controller issues alarms when the distance between two consecutive vehicles is lower than a given limit. We propose to model the controller by Fluid Stochastic Petri Net (FSPN). Several performance indices are evaluated in normal and abnormal conditions.

In the Electronic Funds Transfer (EFT) Systems, faults can cause severe degradation on the performance of this system. Thus, modelling the performance of EFT system without considering dependability aspects can cause inaccurate results. This paper presents a stochastic model for evaluating performance of processing and storage infrastructures of the EFT system. This work also presents a model f...

This paper introduces an analytical method for approximating the performability of a firm realtime system modeled by a multi-server queue. The service discipline in the queue is earliestdeadline- first (EDF), which is an optimal scheduling algorithm. Real-time jobs with exponentially distributed relative deadlines arrive according to a Poisson process. All jobs have deadlines until the end of s...

This paper describes how to construct complex performability models in the context of the software tool Moebius, by hierarchically composing small submodels. In addition to Moebius’ “Join” operator, a second composition operator “Sync” is introduced, and it is shown how both types of composition can be realised on the basis of symbolic, i.e. BDD-based data structures.

Purpose: Describe a method for determining the performability of computer networks under transient fault conditions. Special math needed for explanation: None Special math needed to use results: None Results useful to: Visualize and understand the eeects of transient faults on two carrier sense multiple access protocols.

In this paper we propose a methodology for the modelling, verification and performance evaluation of communication components of software for enterprise information systems. The methodology is centered upon model-driven development using a subset of UML 2.0 diagrams. It is supported by the proSPEX model processing tool which offers a simulation-based executable verification environment. The mod...

We review high-level speciication formalisms for Markovian performability models, thereby emphasising the role of structuring concepts as realised par excellence by stochastic process algebras. Symbolic representations based on decision diagrams are presented, and it is shown that they quite ideally support compositional model construction and analysis.